#chatbot hype party!!!

[seriously-mike]

What Will Destroy AI Image Generation In Two Years?

You are probably deluding yourself that the answer is some miraculous program that will "stop your art from being stolen" or "destroy the plagiarism engines from within". Well...

NOPE.

I can call such an idea stupid, imbecilic, delusional, ignorant, coprolithically idiotic and/or Plain Fucking Dumb. The thing that will destroy image generation, or more precisely, get the generators shut down is simple and really fucking obvious: it's lack of interest.

Tell me: how many articles about "AI art" have you seen in the media in the last two to three months? How many of them actually hyped the thing and weren't covering lawsuits against Midjourney, OpenAI/Microsoft and/or Stability AI? My guess is zilch. Zero. Fuckin' nada. If anything, people are tired of lame, half-assed if not outright insulting pictures posted by the dozen. The hype is dead. Not even the morons from the corner office are buying it. The magical machine that could replace highly-paid artists doesn't exist, and some desperate hucksters are trying to flog topically relevant AI-generated shots on stock image sites at rock-bottom prices in order to wring any money from prospective suckers. This leads us to another thing.

Centralized Models Will Keel Over First

Yes, Midjourney and DALL-E 3 will be seriously hurt by the lack of attention. Come on, rub those two brain cells together: those things are blackboxed, centralized, running on powerful and very expensive hardware that cost a lot to put together and costs a lot to keep running. Sure, Microsoft has a version of DALL-E 3 publicly accessible for free, but the intent is to bilk the schmucks for $20 monthly and sell them access to GPT-4 as well... well, until it turned out that GPT-4 attracts more schmucks than the servers can handle, so there's a waiting list for that one.

Midjourney costs half that, but it doesn't have the additional draw of having an overengineered chatbot still generating a lot of hype itself. That and MJ interface itself is coprolithically idiotic as well - it relies on a third-party program to communicate with the user, as if that even makes sense. Also, despite the improvements, there are still things that Midjourney is just incapable of, as opposed to DALL-E 3 or SDXL. For example, legible text. So right now, they're stuck with storage costs for the sheer number of half-assed images people generated over the last year or so and haven't deleted.

The recent popularity of "Disney memes" made using DALL-E 3 proved that Midjourney is going out of fashion, which should make you happy, and drew the ire of Disney, what with the "brand tarnishing" and everything, which should make you happier. So the schmucks are coming in, but they're not paying and pissing the House of Mouse off. This means what? Yes, costs. With nothing to show for it. Runtime, storage space, the works, and nobody's paying for the privilege of using the tech.

Pissing On The Candle While The House Burns

Yep, that's what you're doing by cheering for bullshit programs like Glaze and Nightshade. Time to dust off both of your brain cells and rub them together, because I have a riddle for you:

An open-source, client-side, decentralized image generator is targeted by software intended to disrupt it. Who profits?

The answer is: the competition. Congratulations, you chucklefucks. Even if those programs aren't a deniable hatchet job funded by Midjourney, Microsoft or Adobe, they indirectly help those corporations. As of now, nobody can prove that either Glaze or Nightshade actually work against the training algorithms of Midjourney and DALL-E 3, which are - surprise surprise! - classified, proprietary, blackboxed and not available to the fucking public, "data scientists" among them. And if they did work, you'd witness a massive gavel brought down on the whole project, DMCA and similar corporation-protecting copygrift bullshit like accusations of reverse-engineering classified and proprietary software included. Just SLAM! and no Glaze, no Nightshade, no nothing. Keep the lawsuit going until the "data scientists" go broke or give up.

Yep, keep rubbing those brain cells together, I'm not done yet. Stable Diffusion can be run on your own computer, without internet access, as long as you have a data model. You don't need a data center, you don't need a server stack with industrial crypto mining hardware installed, a four-year-old gaming computer will do. You don't pay any fees either. And that's what the corporations who have to pay for their permanently besieged high-cost hardware don't like.

And the data models? You can download them for free. Even if the publicly available websites hosting them go under for some reason, you'll probably be able to torrent them or download them from Mega. You don't need to pay for that either, much to the corporations' dismay.

Also, in case you didn't notice, there's one more problem with the generators scraping everything off the Internet willy-nilly:

AI Is Eating Its Own Shit

You probably heard about "data pollution", or the data models coming apart because if they're even partially trained on previously AI-generated images, the background noise they were created from is fucking with the internal workings of the image generators. This is also true of text models, as someone already noticed by having two instances of ChatGPT talk to each other, they devolve into incomprehensible babble. Of course that incident was first met with FUD on one side and joy on the other, because "OMG AI created their own language!" - nope, dementia. Same goes for already-generated images used to train new models: the semantic segmentation subroutines see stuff that is not recognized by humans and even when inspected and having the description supposedly corrected, that noise gets in the way and fucks up the outcome. See? No need to throw another spanner into the machine, because AI does that fine all by itself (as long as it's run by complete morons).

But wait, there's another argument why those bullshit programs are pointless:

They Already Stole Everything

Do you really think someone's gonna steal your new mediocre drawing of a furry gang bang that you probably traced from vintage porno mag scans? They won't, and they don't need to.

For the last several months, even the basement nerds that keep Stable Diffusion going are merely crossbreeding the old data models, because it's faster. How much data are Midjourney and OpenAI sitting on? I don't exactly know, but my very scientific guess is, a shitload, and they nicked it all a year or two ago anyway.

The amount of raw data means jack shit in relation to how well the generator works. Hell, if you saw the monstrosities spewed forth by StabilityAI LAION default models for Stable Diffusion, that's the best proof: basement nerds had to cut down on the amount of data included in their models, sort the images, edit the automatically generated descriptions to be more precise and/or correct in the first place and introduce some stylistic coherence so the whole thing doesn't go off the rails.

And that doesn't change the fact that the development methodology behind the whole thing, proprietary or open-source, is still "make a large enough hammer". It's brute force and will be until it stops being financially viable. When will it stop being financially viable? When people get bored of getting the same kind of repetitive pedestrian shit over and over. And that means soon. Get real for a moment: the data models contain da Vinci, Rembrandt, van Gogh, and that means jack shit. Any concept you ask for will be technically correct at best, but hardly coherent or well thought-out. You'll get pablum. Sanitized if you're using the centralized corporate models, maybe a little more horny if you're running Stable Diffusion with something trained on porn. But whatever falls out of the machine can't compete with art, for reasons.

[samueldays]

AI Winter is coming

mostly @northshorewave who has been worried about this

The other day a friend wanted to show off his positive interactions with ChatGPT. He’d used it to write a 20 questions quiz for the party we were at. I asked if I could proofread the questions and promised not to look up the answers. It contained stuff like:

“Which one of these countries does not have a square flag? a) Switzerland b) China c) Britain d) Egypt”

Switzerland is the only one of them that does have a square flag. ChatGPT is an overgrown autocomplete that “knows” to associate the concepts “square flag” and “Switzerland” and the general shape of a quiz question, and mashes words into this template. When I pointed this out, my friend was rather disappointed and quickly set to reviewing the quiz.

Then I asked to use his laptop for a moment, and showed off ChatGPT’s propensity to hallucinate by asking it for a summary of a nonexistent Wikipedia article whose title I made up on the spot.

ChatGPT happily summarized the article as describing a Danish children’s TV series involving two boys who go to Mars in their home-built spaceship to look for their dad who disappeared on a research expedition there. The series had 12 episodes that ran from 2005 to 2006. It was produced by (Danish studio name I don’t remember) for a cost of fourteen million dollars.

My friend spent the next several minutes on Google checking whether this series had at any point existed, and rushing through the five stages of grief. :^D

This looks like to me like a miniature of the current hype cycle (”AI Summer”), which will die in a year or two, and there will be another “AI Winter” of disappointment. I say another because there’s been at least three and possibly more minor ones. Experts dispute the count, there’s no objective number, but this is my impression:

In the 60s, there was an AI hype cycle. It produced a lot of obscure tech and the moderately famous ELIZA, an early chatbot-psychotherapist. ELIZA arguably passed the Turing test in the very narrow sense of “some people talking to it thought it was human”. AI researchers were sure that full humanlike AI was probably just a decade away.

Enter the 70s, humanlike AI is nowhere close, ELIZA is clearly just a trivial grammar engine, AI winter sets in, people and funding leave the field.

A second hype cycle around “expert systems” AI packed full of knowledge and rules and heuristics started in the early 80s. Surely this time AI is close, now that it knows stuff. Nope - AI winter 2 around late 80s-1990. My pet nerd example here is Eurisko winning a tournament of Traveller TCS - a very very large wide-open sandbox game with custom-designed ships in space battles, which has too many possibilities and too much rock-paper-scissors to solve. Eurisko exploited enough edge cases and loopholes and cheese tactics that the judges changed the rules for next year. Eurisko won the next year again with new cheese, and the judges said “please don’t come back”.

Expert systems did spread into businesses and automation, but nobody thinks “AI” anymore about the automatic crop-picking robot that can tell green crops from green leaves.

In the late 90s, another AI hype cycle starts. A focus for this one is Deep Blue defeating world chess champion Garry Kasparov in 1997. Chess was a proverbial smart people game; is the AI finally smart enough to be humanlike now that it can beat the smartest person in the world?

No. Instead, chess stopped being a proverbial smart people game, now that it could be brute-forced, and Deep Blue looked less smart and more fast, having enough computing power to examine 200 million moves per second by mostly brute force. Well, I’d play a lot better too if I got enough subjective time to examine possible moves on 200 million virtual boards.

AI winter 3 in the 2000s.

Now it’s machine learning that is in another hype cycle around neural networks and machine learning from the mid-2010s (maybe from AlphaGo v Lee Sedol?) and into present-day ChatGPT. Maybe now it’s finally going to be a real boy...

...but history suggests not. It seems likely to me that ChatGPT’s failures will become more blatant, chatbot detection will become more common, and massive disappointment will set in within a few years now that the hype is so high. A few years after AI winter 4, we’ll be accustomed to the rather more limited things that GPT makes a good tool for.

A lot of people are saying that ChatGPT or Bing Sydney passes such-and-such test. Consider: is this a test for which answer material is available on the internet? Because a lot of ChatGPT behavior involves, basically, searching for answers to copypaste in internet-derived training data. This is a great technique for sounding moderately intelligent on any sort of test or in any field; and a terrible technique for advancing the state of the art, or saying anything I couldn’t find with my own search, or showing one’s skill at anything but copypaste.

ChatGPT is like a cheating D student, and its likely applications are on the order of “What if you had infinite D students as unpaid interns?”

(spamming publishing houses with D-grade schlock being one such)

[perfectiongeeks]

ChatGPT Replace Your Dot Net Development Company

Do Dot.NET Developers Need New Skills to Succeed in an AI-Based Industry? Is ChatGPT here to revolutionize the dot net development industry? Experts believe that ChatGPT is here to revolutionize app development. From starting human-like conversations, solving privacy problems, and even writing code for programmers, OpenAI’s new tools have the power and potential to change developers’ jobs. But as the best asp.net development company in Delhi, you want to know what we think about ChatGPT.

The Hype About ChatGPT

ChatGPT was released in November 2022, more than 5 million people have registered for this free chat version, which is based on the GPT-3 family of Open AI (Generative Pretrained Transformer). Chatbot can interact with programmers as a friend and write simple web pages and programming code in JavaScript, Python and React languages. It can also find bugs and errors in code and help you create new programming languages.

ChatGPT was released in November 2022, more than 5 million people have registered for this free chat version, which is based on the GPT-3 family of Open AI (Generative Pretrained Transformer). Chatbot can interact with programmers as a friend and write simple web pages and programming code in JavaScript, Python and React languages. It can also find bugs and errors in code and help you create new programming languages. She also writes poems and songs, and suggests birthday party ideas. We strongly believe that if ChatGPT continues to do what it is doing now, it will one day replace Google. Open AI tools interact with users in a conversational way, answering questions based on past interactions, asking inappropriate questions, discussing topics, and even admitting mistakes. We believe that AI applications are amazing.

What is the Best Way to Let ChatGPT Take Over Programming?

ChatGPT is a brand-new artificial intelligence technology that can create code from natural language descriptions. This technology could eventually eliminate the need for a programmer to write code since ChatGPT could do it for them.

The primary benefit of ChatGPT can make programming easier for programmers by saving them a significant amount of time.Instead of writing codes, they can simply write down what they want to achieve with the code, and ChatGPT will handle the remainder.

This can dramatically accelerate the development process, making it simpler for non-programmers to develop software.

Another advantage could be that ChatGPT could develop better software than humans. For example, it has been demonstrated to be superior to human programmers at tasks like solving bugs.

This means it is possible to create more efficient and reliable software.

There are some drawbacks to ChatGPT. ChatGPT. One of these is that it’s still in the early stages of development, which means there will be some issues and bugs in ChatGPT.

Another reason is that they may not be able to comprehend complicated concepts or perform imaginative work as humans.

In the end, ChatGPT has the potential to replace programming in many instances. In certain situations, it could save time while producing higher-quality code than humans. But it’s still in its early days for this technology, and its capabilities have a few limitations.

[tastydregs]

GOP Launches the Age of AI-Generated Attack Ads

“What if the weakest president we’ve ever had were re-elected?” the video poses right off the bat. Apparently we’d see a lot more cliche AI-generated art. Republicans have waged a full-on attack against President Joe Biden—who just announced his re-election campaign for 2024 today—with a scrapbook video of AI-generated images portraying a bleak portrait of America under Biden’s rule.

Maybe AI-Written Scripts are a Bad Idea?

The video, titled “Beat Biden,” was released on the Republican National Committee’s official YouTube account this morning, shortly after Biden announced his re-election campaign for next year’s presidential race. The video illustrates just how bad the GOP thinks America could get under Biden’s thumb using all AI-generated imagery. Despite the best efforts from the country’s best and brightest Republicans (who are few and far between), the images in “Beat Biden” resemble less of a scathing political critique, and more like something ripped from the movie adaption of a typical young adult dystopian novel. Selected works include sensationalized images of China invading Taiwan, the economy collapsing, immigrants flocking across America’s borders, and a rising crime and opioid crisis.

Beat Biden

The GOP didn’t explicitly mention what AI platform was used to generate the images seen in the video, and present nothing more than “Built entirely with AI imagery” to serve as a disclaimer in the upper left-hand corner of the advertisement. As The Verge reports, the party probably didn’t use Midjourney or DALL-E 2, as those AI generators attempt to limit politicized outputs and prompts. It’s very possible that the GOP used Stable Diffusion, a DALL-E 2 competitor that lets users create any images they want—yes, that includes inflammatory political content such as “Beat Biden.”

While “Beat Biden” is a hare-brained attempt to drum up opposition against the incumbent candidate, it does point to a growing issue with the AI hype train’s place in politics. Prior to the arrest of Truth Social founder Donald Trump earlier this month, AI-generated images of cops aggressively detaining him began to spread across the Internet. If you looked close enough, it was clear that the images were fake, since they have that signature soft, air-brushed quality so much AI imagery has. However it’s not a stretch to think that an untrained eye could believe that these images are real, especially in the emotional anticipation of a former president being arrested.

Shortly after the Trump arrest hoax photos went viral, the platform used to generate those images, Midjourney, halted free trials citing “extraordinary demand and trial abuse.” Midjourney told Gizmodo at the time that shutting off access to the generator had nothing to do with the arrest photos—or the simultaneous photos of the Pope rocking Balenciaga—but was instead a result of people making throwaway accounts to abuse the free trial. Regardless, the Beat Biden, Trump arrest, and hypebeast Pope incidents illustrate a clear need for AI generators to do something to signal what’s real and what’s not—especially if the upcoming 2024 election will be as contentious as 2020 and 2016.

Want to know more about AI, chatbots, and the future of machine learning? Check out our full coverage of artificial intelligence, or browse our guides to The Best Free AI Art Generators, The Best ChatGPT Alternatives, and Everything We Know About OpenAI’s ChatGPT.

[elegantlightparadise]

Why Your Business Needs a Chatbot to Stay Ahead in the Customer-First Era

Are you tired of hearing the same old story from your customers about how they had a terrible experience with your business? It's time to change that tune, and the solution is simple - a chatbot!

In today's customer-first era, businesses can't afford to fall behind. With the rise of technology and instant gratification, customers expect quick and efficient service. And that's where chatbots come in to save the day.

Don't believe the hype? Let me break it down for you. Chatbots can handle a multitude of tasks, from answering basic FAQs to handling customer complaints, all in real-time. Not only does this free up your employees to tackle more complex issues, but it also ensures that your customers always have someone available to assist them, even outside of business hours.

Think about it. How many times have you gotten frustrated when you couldn't get in touch with a company when you needed help the most? With a chatbot, those days are gone! Customers can quickly and easily get the information they need without having to wait on hold for hours or send countless emails back and forth.

But chatbots aren't just about making life easier for customers. They can also provide valuable insights for your business. By tracking and analyzing customer interactions, chatbots can help identify trends and pain points, allowing you to make informed decisions about how to improve the customer experience.

And the cherry on top? Chatbots are cost-effective. No need to worry about hiring a team of customer service reps or paying overtime for after-hours support. Chatbots can handle all of that and more, 24/7, all while keeping your overhead costs low.

So, what are you waiting for? The future is here, and it's time to hop on the chatbot train. Trust me, your customers (and your bottom line) will thank you. Don't be the last one to the party - start reaping the benefits of chatbots today!

>>> Click here to hire the best group of Chatbot developers ,designers and researchers

#Chatbot, #Artificial Intelligence, #Machine Learning,

#Deep Learning App or Services, # Data Science, #Big Data, #Analytics Services

[diinfotechin]

Will ChatGPT Replace Your Dot NET Development Company Programmers?

Do Dot.NET developers need new skills to succeed in the AI-based industry? Is ChatGPT here to change the dynamics of a dot net development company?

Experts believe ChatGPT is here to transform application development. By initiating human-like conversations, solving coding problems, and even writing codes for programmers, OpenAI’s new tool has the capacity and capability to change the careers of developers.

But as the best asp.net development company in Delhi, do you want to know what we think on ChatGPT.

The Hype About ChatGPT

ChatGPT was released in November 2022, and more than 5 million people have registered on this free language model chatbot, based on Open AI’s GPT-3 family (Generative Pretrained Transformer). The chatbot can interact with programmers as a friend and write simple webpages and programming codes in JavaScript, Python, and React languages. It can also find errors and bugs in codes and help you create new programming languages.

It also writes songs and poems, and suggests birthday party ideas.

We are quite positive that if ChatGPT continues to do what it currently does, it will replace Google someday. The Open AI tool interacts with users in a conversational way, answers questions based on previous interactions, reject inappropriate queries, discusses a subject, and even admits its mistake.

We agree that the AI tool is surprisingly great.

will ChatGPT replace your creative and valuable dot net Programmers?

Our answer is No. However…

While we don’t know what will happen 10 years from now, one thing that we can predict is that programmers will no longer need to write boilerplates. They can use their skills in more intensive development areas that AI cannot handle yet, such as cybersecurity.

The work of a programmer from an ASP.NET development company in Delhi is not simply writing codes. They also have to turn ideas into lasting solutions. An AI cannot imitate the million-dollar idea you have in your mind in the same manner.

Simply put, it takes skills to be a programmer – understand customer needs, coordinate with the designer, structure a program, and produce something greater than anticipated.

And ChatGPT doesn’t have all the answers yet. But the good thing is, that it is honest about its shortcomings.

Our Take on ChatGPT vs Developers Skill Is?

While ChatGPT can replace some aspects of programming, such as coding generic functions, it won’t replace Dot NET Development Company Programmers .

However, ChatGPT will pave the way for new jobs in the IT industry. We will see an increase in demand for prompt engineers, AI coders, and software developers specialized in data science principles.

So, the AI era and the AI apps like ChatGPT will only simplify programmers’ jobs and create new job titles and skill sets.

ASP.NET is one of the best open-source frameworks for building modern web applications for Windows, macOS, Docker, and Linux. And as a .NET company in Delhi, we assure you that no AI stimuli can replace our services and solutions.

As a .NET development company developer in Delhi, we see it as a copilot that can be useful in performing repetitive tasks and for making informed decisions.

So, if you are bothered about the viral articles, memes, and clickbait videos about how ChatGPT will replace Dot NET Development Company programmers and content creators’ jobs, don’t worry. It is all but hype.

[jaeyooniverse]

Hello! Fantasy friend here 💕

I don't really have plans for my birthday, I'll probably just spend it with my family - I honestly prefer not to make a big fuss over things like my birthday, as long as I have people I love around it's enough for me ☺️

Omg yeah kids/teens online... it's terrifying to me how much kids share on the internet and how much they absorb, there are so many awful people and people who encourage unhealthy/toxic mindsets on here and it's basically impossible to stay away from influences like that especially if you're not already versed in how to deal with people and know what kinds of things are ok to say and do. Like, the most we can do personally is to try and look out for kids online and be good influences to them, but man sometimes it's hard not to just be like "you are Too Young for the internet go play in your backyard or something it's too dangerous for you on here"

Computational linguistics sounds really cool! From the sound of it I would assume that it's an area of linguistics that has something to do with math, but I could be super wrong - what does it involve?

Oh I bet Puerto Rico and Nigeria were both super awesome to visit! California is pretty cool too (I live in California so I have to hype it up lol), just make sure to come when it's not on fire 😂

Hope your day goes well! 💖

Oh that still sounds nice! Yeah I usually don't have big celebrations for my birthday I usually just get something nice to eat and hang out with friends and family. I think going for a little vacation could be nice? Like for my bday a few years ago I went out of state with a friend and we saw DAY6 and just had fun doing whatever. It was fun and outside of the concert it was still a pretty lowkey chill weekend. But I would never really want to have like a huuuge party with a bunch of people. I already feel awkward enough having small celebrations for me so I'd rather not 😂

Exactly 😭😭 Like luckily sometimes, as easy they pick up on the bad stuff, if they find that positive influence they're able to unlearn it just as well. But yeah sometimes I'm like please just log off,, find some toys to play with, read a book, run around outside with your friends while you have the energy and time to do that 😂 Seeing people be like they were little/babies when bts or someone debuted like...okay you shouldnt be on twitter rn how did you get here 😭

ahaha doesn't it sound cool? I'm still learning it I'm fairly new to it since in undergrad I studied math. Computational Linguistics is more on the technology/computer science side of linguistics. There a lot of different applications but for example it deals with language processing like translators, speech-to-text, chatbots, and search engines. I'm primarily interested in learning about AI/chatbots but after my first year I've been introduced to other parts of the linguistics field that I've found myself interested in, like semantics (primarily bc we discussed semantics using functions and logic, which, as somewhat of a math-enthusiast, I find pretty cool!)

Nigeria was way back like...16 years ago lmao I don't remember a lot of what we did but I do remember enjoying it and seeing my family. The Puerto Rico trip was fun but too short 😭😭 There were so many things my friend and I wanted to do but we were only there for a couple days. We went for a nature walk in a rainforest it was so beautiful :')

dhfsajkhk I hope you're doing okay! I have a couple mutuals in cali and they've been telling me about the heat...idk how you all are surviving and I heard it's only gonna get hotter?? 😭 whenever it cools down let me know then i'll head over 😂😂

And okay you say you have to hype it up bc you live there but honestly I don't hype up my state like that 😂😂 maybe it's bc I haven't done all the fun stuff it has to offer but I mean I've been to DC and that's pretty much where tourists always go when they come here 😂

I hope you're having a nice weekend ^^

[pyramidskilltech]

AI Today: Who Is Using It Right Now, and How

Takeaway: AI is a versatile tool, but how is it currently being used in business? Here we take a look at some of the implementations.

Artificial intelligence is all the rage in the enterprise these days. Stories abound about all the gee-whiz capabilities it will bring to our personal and professional lives.

But like any technology, there is usually a fair amount of hype before the reality sets in. So at this point, it is probably worth asking: Who is using AI right now, and how?

AI in Action

In a broad sense, says Information Age’s Nick Ismail, AI is already bringing five key capabilities to the enterprise:

AI in Action

In a broad sense, says Information Age’s Nick Ismail, AI is already bringing five key capabilities to the enterprise:

        Voice/Image     Recognition: Applications range from accurately transcribing meetings and     sales calls to researching the impact of branding, logos and other visuals     on the web.        

Data     Analysis: Unstructured data in particular is very difficult to     quantify. Using readily available tools, organizations are able to delve     into the minutia of their operations, supply chains, customer relations     and a wealth of other activities to gather intelligence that is both     accurate and actionable.        

Language     Translation: Convert one spoken language into another in real time, an     increasingly important tool for multi-national corporations.        

Chatbots: Automate the customer     experience with a friendly, responsive assistant that can intuitively     direct inquires to the proper knowledge base.        

Predictive     Analysis: Accurately forecast key data trends, such as cash flows,     customer demand and pricing.

To see some of these capabilities in action, check out the new website for Peach Aviation which features an automated response system that provides multi-language support for customer inquiries. The system runs on the Desse AI agent provided by SCSK ServiceWare Corp., and can respond in all languages serviced by the airline: Japanese, English, traditional and simplified Chinese, Cantonese, Korean and Thai. As well, it uses data analysis to continuously monitor questions and answers to provide steadily improved quality. The company reports that out of 100,000 inquiries received in late December and early January, the system was able to provide automatic responses to 87 percent.

Yet another example of AI in action is a joint project by NBCUniversal and CognitiveScale to discern the key elements in a successful Super Bowl ad. The companies used CognitiveScale’s Cortex platform to analyze three years’ worth of game-day commercials and various client-engagement data to derive actionable insights linked to key video concepts, attributes and themes. For instance, the research showed that comedic effects work best with sales messages, while uplifting tones are more effective for branding.

While AI will not write and produce the perfect ad itself, NBCUniversal’s SVP of Corporate Analytics and Strategy Cameron Davies said it provides greater insight into what works and what doesn’t. “The CognitiveScale platform gives us the ability to consider new ad strategies for companies who want to ensure their ads will be successful when they invest in production and media buying,” he said.

CognitiveScale is also working with organizations in the financial, health care and retail industries by allowing video data to undergo the same analytics processes as voice, image and text.

Baddies Beware

AI is also turning into an effective crime-fighting tool, says Forbes’ Rebecca Sadwick. It turns out that one of the biggest hindrances to modern law enforcement is the bureaucratic inertia that exists in both public and private processes. AI helps overcome these hurdles, bringing much-needed clarity to highly organized criminal enterprises ranging from money laundering to human trafficking to terrorism.

One of the key ways AI helps solve crimes is by lowering the cost on private entities to oversee their transactions. Like any regulatory requirement, compliance is primarily a cost factor for organizations that are focused on profitability. Using third-party AI platforms specifically geared toward identifying suspicious data patterns, companies have not only lowered their costs but increased their chances of detecting nefarious activities. Prior to AI, it is estimated that nearly half of all financial crimes went unnoticed.

As well, banks and financial institutions that have deployed AI in this way actually help law-abiding citizens take part in fighting crime. Every time a legal transaction is processed, a learning  is exposed to the normal patterns of money movement and is thus better equipped to identify transactions that break these patterns.

Technology is a two-way street, of course, so the same technology that is currently helping to fight crime can also be used to conduct it. With enough computing power, an intelligent system might be able to leverage the rising trend of micro-transactions by breaking up large transactions into numerous, perhaps millions, of smaller ones that are harder to detect and track. As well, quantum technology has yet to make its presence known in the criminal underworld (as far as we know, at least), which would open up an entirely new front in the war against cyber crime.

Clearly, we are in the earliest stages of AI development, so there will no doubt be numerous other ways in which it will affect mainstream enterprise processes as the market matures. Unlike earlier technologies, however, AI is expected to improve with age as it incorporates both human- and machine-generated data to forge a greater understanding of the environment it occupies and how best to navigate it.

And this is likely to be the most profound change of all: the end of lengthy development processes in which new features come out once a year (if that) and can only be implemented by taking infrastructure and data offline. In the future, digital systems will get better with age, all by themselves.

[warmdevs]

New Post has been published on https://warmdevs.com/wechat-chinas-integrated-internet-user-experience.html

WeChat: China’s Integrated Internet User Experience

The big Internet companies of the Western world have proclaimed “conversational user interfaces” and chatbots to be the Next Big Thing, with major initiatives led by Amazon, Apple, Facebook, Google, and Microsoft. Even Taco Bell has a “TacoBot,” a service that allows users to order and customize their tacos using a natural-language UI.

Much of this hype stems from angst generated by the success of the Chinese WeChat service, which had 700 million users as of April 2016. WeChat has been touted as the poster child for conversational user interfaces. In this article, we report on user research we did in China with WeChat users. The study aimed to uncover practices in which WeChat users engaged, as well as why and in which cases its users preferred to use WeChat instead of regular mobile websites and apps.

WeChat: All You Want in One Place

As the name implies, WeChat started as a mobile instant-messaging (IM) app, launched by Tencent in 2011. However, WeChat quickly added other features; today, it is no longer primarily an IM application, but it also includes:

A payment service. 200 million users have connected their bank accounts to WeChat. This capability allows person-to-person money transfer, including 8 billion “red packets” (so called because gifts of lucky money were traditionally given in red envelopes) sent during the 2016 Chinese New Year celebrations, as well as easy payments to either online or real-world businesses.

A platform for companies’ online presence, with 10 million “official accounts.” Official accounts are WeChat-supported mini-websites that are easy to establish and especially convenient for small businesses, compared with the high-tech skills needed to build and maintain an interactive website.

An ecommerce platform, which is particularly convenient because it’s integrated with the payment service.

Social networking. The “Moments” feature is somewhat like a Facebook wall where friends can post messages in different formats (e.g., text, audio, pictures, stickers, videos). Furthermore, message dialogues allow users to interchange text, audio, pictures, stickers, short videos that are taken within WeChat, links, contact cards, and documents.

Other social services, such as “people nearby.”

Games.

A broad range of 3rd party services. Examples include ordering a taxi, booking train and airline tickets, buying movie tickets, and paying utility bills.

Integration with the physical world. QR scanning is widely used to quickly access WeChat official accounts, and exchange information (for example, contact information). There are even plush toys that connect to WeChat.

The Mon Mon plush toy for little children to interchange voice messages with their parents’ WeChat accounts. (Still from advertisement created by the vendor, the Chinese toy company Dan Dan Man.)

In total, more than a third of the time spent by Chinese users on mobile is on WeChat. No wonder Western companies want to get in on the action.

However, the key UX advantage of WeChat is not that it grew out of a chat service; it’s the integrated user experience. Each individual service is fine, but not necessarily better than those offered by other companies. In fact, our user testing of WeChat revealed many usability problems in various areas. What’s superior is how these services play together and reinforce each other. Most importantly, these benefits are not the result of a superior, simple conversational UI; instead, they are often provided through a simplified graphical user interface (GUI).

WeChat is similar with a megaportal providing access to a huge number of services. It’s almost a parallel, alternate web, whose explosive evolution seems to be rooted in two different aspects:

A huge initial adoption of the IM service and, later, of the payment ability

The scarcity of mobile-optimized sites and services on the Chinese web

The first aspect led to widespread adoption and provided a convenient substitute for credit cards (a less common method of payment in China); the second provided one (and often the only) convenient way of online access to a business or a service.

We asked our Chinese research participants to draw their mental models of WeChat, and as shown in the following examples, many of them considered the smooth payment service and the official accounts more central than that the chat service.

Mental models of WeChat, as drawn by two of our study participants.

User Research

We conducted two rounds of user research in China:

A diary study, in which 12 participants logged their WeChat use for 7 days:

Users sent us a WeChat message each time they used a special WeChat functionality.

Users also completed two questionnaires at the end of each day: one about their use of basic WeChat features and one about their use of special WeChat features. Basic features were defined as messages and chat with friends, checking Moments (the Facebook-wall equivalent), and reading articles published by official accounts. All other activities were considered special features; these included interacting with official accounts, posting to Moments, using payments, scanning QR codes, and shopping.

A usability test, using the thinking-aloud method, where 14 participants used WeChat and other mobile apps to perform representative tasks during 90-minute sessions. At the end of these sessions, most users sketched their mental models of WeChat.

About half of the users were in Beijing (one of China’s advanced “tier-1” cities, with 21 million inhabitants) and the other half were in Tangshan (a smaller and more traditional industrial “tier-3” city, with 7 million inhabitants). The specific location doesn’t usually matter for usability testing, but, in this case, market research suggests that WeChat penetration varies in different parts of China: 93% market penetration in tier-1 cities vs. 43% in tier-3 cities. Local services (e.g., offline businesses using WeChat official accounts and payment tools) also differ in these two types of cities.

About half the study participants owned iPhones and the other half owned Android phones.

Here are some examples of usability-testing tasks (translated into English):

You are planning a trip to Beidaihe next Tuesday. Use WeChat to find what train tickets are available.

Use WeChat to check how crowded it is now at Sanlitun.

Your friend recommended “Jacky’s Crawfish.” Order some crawfish, and have it delivered to our testing location. Please stop before just before completing the purchase.

WeChat Payments Rule

In the diary study, fully 32% of all recorded WeChat activities were payments, confirming this service’s central role in WeChat. On average, during the 7-day diary, each participant made 6 payments — about one per day. The majority of the payments were directed towards offline businesses or physical persons, and actual ecommerce payments were in the minority.

Recipients of WeChat payments made by our study participants.

The main reason for the popularity of the WeChat payment service was its smooth user experience and its ability to cut across channels and across the business/social divide. As one study participant said, “When I am paying with WeChat, I only need to take my phone out — no bank card, no signatures, no cash, no change — so convenient!”

The next two major uses of WeChat were Moments (the Facebook equivalent) with 16% of use and official accounts (miniwebsites hosted by WeChat) with 10% of use. The many other WeChat features accounted for the remaining interactions and thus summed to 42% percent of the overall use.

Integration and Consistency of User Experience

What makes WeChat superior is its integration: many features that all work together and build on each other. Once users subscribe to an official account, they can receive content pushed by that account. They can also access the WeChat page of that account and find more relevant information about the company or read associated pieces of content. They can purchase products or send money to the company. Thus, subscribing to an account opens up a world of interaction possibilities with that company, all strongly tied into WeChat.

Palace Museum account: The user can select one of the tabs at the bottom of the screen to interact with the Palace Museum (top left). Each tab opens up a submenu (top middle). A list of articles about the Palace Museum accessed through one of the menu options (top right). A WeChat “webpage” for the Palace Museum can be accessed through another one of the menus (bottom). (Note, however, that the Palace Museum’s webpage accessed through a regular mobile browser is actually not mobile optimized.)

These WeChat services are often not available through other mobile channels — for example, the Palace Museum does not have a mobile-optimized website available through a mobile browser. Thus, many users will prefer to interact with a company through the consistent and relatively predictable WeChat interface instead of risking to go to a different channel and have an inferior user experience.

The Palace Museum website as accessed through a mobile browser is not mobile optimized.

For companies, WeChat is an inexpensive way to ensure a presence on mobile; that presence has the advantage of offering a standardized, consistent user experience. Users are likely to succeed when engaging with an official account because most of them look the same — they are the equivalent of websites designed according to the same design pattern. The interaction is thus easy to learn and is made of the same WeChat supported building blocks that people are already familiar with from countless previous interactions with other companies.

WeChat vs. Traditional Websites

Although the mobile web is perhaps not as strong in China at this point (with many sites not having mobile versions), it is most likely that WeChat and regular websites will continue to coexist because they each have advantages and disadvantages. Our users didn’t expect the WeChat official accounts to cover all their needs, but they saw them as an easy way to access commonly used features.

Our users said that they preferred using companies’ official accounts on WeChat for several reasons:

Ease of access. WeChat is the primary QR-scanning application, freeing users from typing which has high interaction cost on mobile phones and is particularly unpleasant for older users.

Ease of interaction. As discussed above, it is easier to interact with an official account — you know where to click and how to use it because they all share the same basic interaction style.

Access to small businesses. Many such business (e.g., a fruit shop, handcraft, interest-based organizations) are unlikely to invest in a real website.

Special discounts. Followers of a WeChat account often receive promotions or exclusive campaigns.

Up-to-date information. There’s obviously no fundamental reason why a company’s website could not be kept as current as an official account, but users’ perception was that websites were more static, whereas WeChat accounts were updated more frequently.

While WeChat official accounts have their benefits, users continued to use traditional websites in parallel with WeChat, for several reasons:

More complex functionality, beyond the simple WeChat-supported UI

Spam: Too many articles pushed by WeChat official accounts can easily become annoying, even for followers of those accounts. Some abuse users’ trust and flood them with ads.

Closed system: WeChat excludes access to frequently used services such as Taobao (a major ecommerce site) and important music providers.

Impaired findability: Often different accounts will have similar names, and some of them may belong to a legit company and others may be fake and trying to impersonate that company. To make things worse, the same company may have different WeChat accounts, and the accounts themselves may be of different types (service accounts vs. subscription accounts) with different capabilities (service accounts will support payments, while subscription accounts won’t). Finding the right account can thus be challenging for users.

Trust issues: Some articles on WeChat spread fake information.

WeChat search results for Palace Museum (left) and for Xuxian, the fruit business (right). Checkmarks indicate that the account is verified by WeChat. But these verification checkmarks do not help identify the desired account, because many verified results may actually be returned. Some of these results will belong to the same company but will have different functionalities, and some will belong to a fake company that may take advantage of a better-known brand’s name. Users can differentiate companies by clicking on their introduction page or checking their messaging history, but these operations are time consuming.

Some of these problems are similar to those faced by users of traditional search engines, whereas others are unique to WeChat, especially the peculiar distinction between subscription and service accounts, which breaks the integrated model of the service and seems to be a legacy of WeChat’s history.

WeChat Usability Problems

Our user testing revealed plenty of usability problems in WeChat and the various official accounts we tested. This is no surprise, since the perfect user interface hasn’t been built yet, but should come as a warning to those technology enthusiasts who believe that simply moving to a new platform or embracing a new interaction style will result in great user experience. On the contrary: any design will still need polishing and adaptation to users’ real needs, as opposed to companies’ internal hopes for what their customers will do. We’ll just give two examples here.

Team Maker allows users to define group events. The setup screen requires users to enter a name for the new event, but 10 of 14 study participants overlooked this field and went straight for the second field (denoted by a clock icon) to enter the time of the event.

Team Maker’s official account: The top field had the instructions “Event Title (Required),” but it was overlooked by most test users.

Several design issues combined to cause this problem:

Banner blindness possibly made users overlook the top field because of the graphic.

The prompt was shown as a placeholder text within the field, which is less noticeable than a separate field label.

That field was not aligned with the other fields in the form, and thus looked as if it was not part of a form, but rather a static title.

The activity title prompt was shown in low-contrast text, especially compared with the brighter text used for the time field.

Users may have had a selfish-action bias, preferring fields related to their own needs (the time of the group meeting) over fields related to system needs (the title).

City Heatmap is a WeChat feature that indicates how crowded a given location was. This feature is available under Wallet → Public Services → City Heatmap. Our users had great difficulty finding it. This navigational path was simply not expected, and suffered from a convoluted information architecture and problematic naming with poor information scent.

Such issues are not special to WeChat and we’ve seen them many times in usability studies. New platforms have old usability issues, because the main determinant of usability is the human mind and people’s needs, not technology.

Text-Based Interaction Gets Limited Use

WeChat supports two different types of interaction with an account:

Text-based: Users can text a number or a keyword to a service account and receive an IM reply from that account. (The reply can be automatic or can be generated by a customer-service representative in charge with responding to WeChat queries).

Menu/link based: Users can select a link embedded in a text message or they can use the menu buttons available on official-account interfaces and in the chat window.

Sometimes all these interaction methods were available for an account, but in other situations only the more basic text interface was supported.

Palace Museum’s official account: users can interact with this account either by texting back one of the numbers 0–6, or by tapping the GUI menu button in the bottom left corner of the screen and selecting some of the available options. On some pages (but not in this case), the numeric options inside the text message are themselves tappable links. (In fact, many of our test participants tried to tap the options inside the message and were disappointed when that action did not produce any result.)

In our diary study and in the usability-testing sessions, we noticed only limited use of the text interface. In particular, some users texted back a number in response to a first message received after subscribing to a service account. They also occasionally texted back a keyword hoping to get back matching search results from that company. However, most users much preferred a menu-based interaction over the more effortful text-based interaction. Whenever the bottom menu was available for a service account, users relied on it, as a more expedient way to interact with the company. They also attempted to tap on the numerical options displayed in the chat message as an even faster way to make a choice.

Our initial interest in WeChat was to better understand conversational text interfaces. The appeal of such natural-language interfaces is that they supposedly allow users to simply express their goal and then sit back, while letting the site do all the work for them. No clicks or taps are involved once the initial query has been formulated, so in a sense, such interfaces have the potential of getting closer to the holy grail of usability ­— zero interaction cost. (Such interfaces do assume that users will be able to formulate a goal — an assumption that does not always hold, because users do not always know the search space well enough.)

We did not find evidence for sophisticated natural-language understanding in WeChat. (Those human-staffed official accounts that reply to individual user queries are definitely not scalable in the long run and were not the norm in our study.) Instead of a true conversational text interface, we discovered a system that warrants the interest of an evolutionary web scientist for the way in which it mimics the evolution of the mobile web — a world in which historic, simple interaction such as the numeric-menu selection and keyword-input coexist with more sophisticated menu-based interfaces or GUIs. These latter methods are newer and often preferred by WeChat users for their lower interaction cost, yet, at this point, they have not yet completely erased the chat-box interface.

Conclusion: WeChat Integration Builds Convenience

If conversational interfaces are not a strength of WeChat, what is the secret behind its success? When we asked our users for their top likes about WeChat, the most frequently mentioned attribute was convenience.

WeChat shines in several ways that build convenience:

A wide variety of services and features

Integration of these features among themselves and, especially in the case of WeChat payment and QR codes, with the physical world

Simple, consistent interaction that stays the same across different official accounts — in marked contrast to the diversity of user interface techniques that plague websites

Integration and simplicity are especially important because WeChat is mainly a mobile service. User interfaces are more difficult to use on mobile devices because of their screen-size limitations and error-prone touch input. A simple interaction style that provides a seamless omnichannel customer journey is essential for mobile designs.

Different usability aspects of WeChat were cited as major positives (second most common after convenience) and as major negatives by our study participants. No UI is completely good or completely bad in terms of usability, and while this interface had many good parts, there were others that definitely needed improvement.

Our study participants’ likes and dislikes about WeChat.

The interplay between usability and convenience points to one of the major contrasts in our field — that between usability and user experience (UX). While usability applies to well-designed user interfaces (UIs), usability is not enough for a good UX. Users need useful features that address their real needs instead of their imaginary ones. When a system is convenient, it goes beyond usability and succeeds in being useful and simplifying a previously complicated task.

WeChat does all of this, and that’s why it’s successful. Not because it grew out of a chat site or originally used a conversational user interface.

[gravitas-ai]

Integrating Artificial Intelligence in your business: A brief guide

Business leaders are keen to integrate an AI strategy to strengthen their sales. In every industry, competitive pressures are raging high today. A powerful strategy based on artificial intelligence can boost up your business growth. It is exciting to learn about the hype around artificial intelligence.

In a global survey involving over 3000 executives, industry analysts and managers and interviews with over 30 tech experts,  85% of business executives were found to vouch for AI. This population believed AI can bring a strategic advantage to their business.

Also, in the last couple of years, AI has gained a significant momentum in industries like healthcare, manufacturing, logistics, retail and transportation. In order to leverage productivity and performance, business firms are integrating AI into their operational mechanism.

Here is a comprehensive guide, which will help you implement AI successfully in your business.

Frame your business strategy

At the outset, business owners need to identify the problems that need to be addressed. Do not jump to the most complex AI implementation. Start with the basics, understanding the technology.

The McKinsey report on AI implementation in 2018 reveals that only in 16% of the cases, a brand new AI innovation is possible. In order to optimize the results, it is important to focus on your long-term strategic goals and implement the strategy accordingly.

Before you finally incorporate the strategy, focus on aspects like:

●       Whether the AI strategy is right for your business

●       Whether it will help you deliver smarter services and products

●       Whether or not your business priorities have changed

Once you are clear on these points, think about the areas where you can use AI to strengthen your business.

Specific AI priorities

Next, it is necessary to focus on the strategic AI priorities. Forward-thinking business firms consider their top priorities and approach the problem accordingly. Depending on the nature of your business, you must think how AI can help you get across to your goals.

Here are certain examples that can give you a better idea:

●       Automating the process of manufacturing

●       Making the sales, HR, accounts and other functions intelligent

●       Automating the mundane or repetitive tasks, so that you can save on manpower

●       Coming up with smarter services and products

Today, a large number of companies are using AI-based analytics in their decision-making process. Regardless of what you are dealing with, you can use these valuable insights to make strategic decisions.

Framing your data strategy

You should know AI works on a lot of data. It is important for you to review the data strategy, so that you can incorporate the AI policy seamlessly. In this regard, you should focus on the following aspects:

●       Where do you source the data from?

●       Do you have adequate amounts of data?

●       In order to get across to the AI priorities, do you have the right kind of data?

●       Will you be using data collected from third parties or set up new methods to collect the same?

●       What are your plans to collect data more strategically?

Once you can stream authentic data into your business, you can ensure a greater accuracy of the insights you gain through AI implementation.

Automate your marketing process

AI-based applications like chatbots can help you automate the marketing process. Most companies implementing AI have been successful in dealing with the routine tasks seamlessly, using the technology.

For instance, you can handle the customer service requests at the initial stages using chatbots backed by AI. It can also help you enhance the overall satisfaction level of the customers and recommend products for upselling.

Through machine learning, you can also benefit in the price-optimization process for all types of markets. Many data science platforms like RapidMiner have been developed, where you can gain information about supplies, competitors, consumer preferences and risk factors.

This information can help you frame your pricing model and target specific segments within the market automatically. Implementing the AI-oriented approach, you can get the profit margins optimized.

Legal and ethical aspects

In this context, you should also focus on the ethical and legal issues. Leading business firms use intelligent machines, which do not interfere with the privacy of people. Find out whether any legal implication exists regarding the usage of AI in the way you intend.

You might be requiring some sort of consent from the users, employees or customers. While you integrate the new technology, do not mess up with the legal and ethical aspects. Another important aspect to consider is whether or not the AI-generated data is free from discrimination and bias.

Bring in the experts

This is perhaps the most crucial stage in the AI implementation process. Leading AI companies are collaborating with business firms, guiding them through a seamless execution process.

When you are ready to implement AI from the technology and organizational standpoint, the most common policy is to set up small and achievable goals and not big and ambitious ones. When you seek a professional support, the experts can deliver invaluable support in the implementation process.

It is necessary to set up a pilot project, before going ahead with your complete strategy. On an average, companies spend 2 to 3 months testing their plans. When you are done with the pilot project, you can think about the more elaborate details for the long-term. Reputed AI experts work in a close collaboration with your in-house people to implement the project.

Enhance your technology, skills and capacity

Once you have implemented the pilot project, you can work on the shortcomings. Identify the skill and technology gap on your platform and acknowledge the same. Seek expert services to bridge this gap and enhance the overall technology infrastructure.

Machine learning, reinforcement learning and deep learning are some of the aspects you need to focus on. You may not have the right technology in place. In this case, have a consultation with the AI experts regarding the systems to realize your goals.

Large business firms have already changed their management policies to implement AI. Think of the teams and employees who will be influenced as a result of AI integration. Communicate with these people about the changes that are likely to occur before implementing the project.

Implementing AI may also have an impact on the company culture. It is wise to think about all these aspects in advance. This ensures you can seamlessly implement artificial intelligence in your business.

[un-enfant-immature]

Rasa raises $13M led by Accel for its developer-friendly open source approach to chatbots

Conversational AI and the use of chatbots have been through multiple cycles of hype and disillusionment in the tech world. You know the story: first you get a launch from the likes of Apple, Facebook, Microsoft, Amazon, Google or any number of other companies, and then you get the many examples of how their services don’t work as intended at the slightest challenge. But time brings improvements and more focused expectations, and today a startup that has been harnessing all those learnings is announcing funding to take its own approach to conversational AI to the next level.

Rasa, which has built an open source platform for third parties to design and manage their own conversational (text or voice) AI chatbots, is today announcing that it has raised $13 million in a Series A round of funding led by Accel, with participation also from Basis Set Ventures, Greg Brockman (Co-founder & CTO OpenAI), Daniel Dines (Founder & CEO UiPath) and Mitchell Hashimoto (Co-founder & CTO Hashicorp). Rasa was founded in Berlin, but with this round, it be moving its headquarters to San Francisco with a plan to hire more people there in sales, marketing and business development; and to continue its tech development with its roadmap including plans to expand the platform to cover images, too.

The company was founded 2.5 years ago, by co-founder/CEO Alex Weidauer’s own admission “when chatbot hype was at its peak.” Rasa itself was not immune to it, too: “Everyone wanted to automate conversations, and so we set out to build something, too,” he said. “But we quickly realised it was extremely hard to do and that the developer tools were just not there yet.”

Rather than posing an insurmountable roadblock, the shortcomings of chatbots became the problem that Rasa set out to fix.

Alan Nichols, the co-founder who is now the CTO, is an AI PhD, but not in natural language as you might expect, but in machine learning. “What we do is more is address this as a mathematical, machine learning problem rather than one of language,” Weidauer said. Specifically, that means building a model that can be used by any company to tap its own resources to train their bots, in particular with unstructured information, which has been one of the trickier problems to solve in conversational AI.

At a time when many have raised concerns about who might “own” the progress of artificial intelligence, and specifically the data that goes into building these systems, Rasa’s approach is a refreshing one.

Typically, when an organization wants to build an AI chatbot either to interact with customers or to run something in the backend of their business, their developers most commonly opt for third-party cloud APIs that have restrictions on how they can be customized, or they build their own from scratch, but if the organization is not already a large tech company, it will be challenged to have the human or other resources to execute this.

Rasa underscores an emerging trend for a strong third contender. The company has built a stack of tools that it has open sourced, meaning that anyone (and thousands of developers do) use it for free, with a paid enterprise version including extra tools including customer support, testing and training tools, and production container deployment. (It’s priced depending on size of organization and usage.)

Importantly, whichever package is used, the tools run on a company’s own training data; and the company can ultimately host their bots wherever they choose, which have been some of the unique selling points for those using Rasa’s platform, when they are less interested in working with organizations that might also be competitors.

Adobe’s new AI assistant for searching on Adobe Stock, which has some 100 million images, was built on Rasa.

“We wanted to give our users an AI assistant that lets them search with natural language commands,” said Brett Butterfield, director of software development at Adobe, in a statement. “We looked at several online services, and, in the end, Rasa was the clear choice because we were able to host our own servers and protect our user’s data privacy. Being able to automate full conversations and the fact it is open source were key elements for us.” Other customers include Parallon and TalkSpace, Zurich and Allianz, Telekom, and UBS.

Open source has become big business in the last several years, and so a startup that’s built an AI platform that has a very direct application in the enterprise built on it presents an an obvious attraction for VCs.

“Automation is the next battleground for the enterprise, and while this is a very difficult space to win, especially for unstructured information like text and voice, we are confident Rasa has what it takes given their impressive adoption by developers,” said Andrei Brasoveanu, partner at Accel, in a statement. “Existing solutions don’t let in-house developer teams control their own automation destiny. Rasa is applying commercial open source software solutions for AI environments similarly to what open source leaders such as Cloudera, Mulesoft, and Hashicorp have done for others.”

[tastydregs]

Robot Lawyers Are About to Flood the Courts

The hype cycle for chatbots—software that can generate convincing strings of words from a simple prompt—is in full swing. Few industries are more panicked than lawyers, who have been investing in tools to generate and process legal documents for years. After all, you might joke, what are lawyers but primitive human chatbots, generating convincing strings of words from simple prompts?

For America’s state and local courts, this joke is about to get a lot less funny, fast. Debt collection agencies are already flooding courts and ambushing ordinary people with thousands of low-quality, small-dollar cases. Courts are woefully unprepared for a future where anyone with a chatbot can become a high-volume filer, or where ordinary people might rely on chatbots for desperately-needed legal advice.

Garbage In, Garnishments Out

When you imagine a court, you might picture two opposing lawyers arguing before a judge, and perhaps a jury. That picture is mostly an illusion. Americans have the right to an attorney only when they’re accused of a crime—for everything else, you’re on your own. As a result, the vast majority of civil cases in state and local courts have at least one party who does not have a lawyer, often because they have no other option. And because court processes are designed for lawyers, every case with a self-represented litigant requires more resources from courts, assuming the person without a lawyer shows up at all. 

Add enough cases like this to a court’s docket, and the results are ugly. In the aftermath of the 2008 financial crisis, thousands of foreclosure cases hit court dockets all at once. Many of the cases were rife with defects: false affidavits, bad notarizations, backdated paperwork, inadequate documentation, and so on. But foreclosures were pushed through anyway, and people lost their homes.

This wasn’t a one-off. It’s a warning of what happens when the world changes and courts don’t adapt. To see that future for robot lawyers, take today’s high-volume filers:  debt collections agencies. Small-dollar ($5,000 or less) debt cases, filed en masse by collections agencies, increasingly dominate local court dockets. While nationwide data is hard to find (more on that later), in 2013, the Pew Charitable Trusts found that small-dollar debt cases made up a quarter of all civil (non-criminal) cases filed in the United States. In 1993, it was just over 10 percent. And cases are on the rise, in red and blue states.

The goal of debt collection cases is simple: Turn hard-to-collect debt into easy-to-collect wage garnishments. In most states, when someone loses a debt case, a court can order their employer to redirect their wages toward a creditor instead. The easiest way for that to happen? When the defendant doesn’t show up, defaulting the case. The majority of debt cases end in default: Either the defendant chooses not to show, is confused about what they need to do or should do, or, just as often, never receives notice of a case at all. “Sewer service,” where plaintiffs deliberately avoid notifying defendants of a legal case (for example, by sending a case to an old address), has been a festering problem in debt and eviction cases for decades, and continues to this day. In some cases, people find out they’ve been sued only after noticing that their paycheck has been garnished.

When a case does default, many courts will simply grant whatever judgment the plaintiff has requested, without checking whether the plaintiff has provided adequate (or any) documentation that the plaintiff owns the debt, that the defendant still owes the debt, or whether the defendant has been properly notified of the case. Sometimes, even the math is wrong: One study of Utah’s courts found that 9.3 percent of debt cases miscalculated the interest plaintiffs were entitled to after a judgment. In other words: garbage in, garnishments out. 

[itunesbooks]

Mastering AWS Lambda - Yohan Wadia & Udita Gupta

Mastering AWS Lambda Yohan Wadia & Udita Gupta Genre: System Administration Price: $35.99 Publish Date: August 11, 2017 Publisher: Packt Publishing Seller: Ingram DV LLC Build cost-effective and highly scalable Serverless applications using AWS Lambda. About This Book • Leverage AWS Lambda to significantly lower your infrastructure costs and deploy out massively scalable, event-driven systems and applications • Learn how to design and build Lambda functions using real-world examples and implementation scenarios • Explore the Serverless ecosystem with a variety of toolsets and AWS services including DynamoDB, API Gateway, and much more! Who This Book Is For If you are a Cloud administrator and/or developer who wishes to explore, learn, and leverage AWS Lambda to design, build, and deploy Serverless applications in the cloud, then this is the book for you! The book assumes you have some prior knowledge and hands-on experience with AWS core services such as EC2, IAM, S3, along with the knowledge to work with any popular programming language such as Node.Js, Java, C#, and so on. What You Will Learn • Understand the hype, significance, and business benefits of Serverless computing and applications • Plunge into the Serverless world of AWS Lambda and master its core components and how it works • Find out how to effectively and efficiently design, develop, and test Lambda functions using Node.js, along with some keen coding insights and best practices • Explore best practices to effectively monitor and troubleshoot Serverless applications using AWS CloudWatch and other third-party services in the form of Datadog and Loggly • Quickly design and develop Serverless applications by leveraging AWS Lambda, DynamoDB, and API Gateway using the Serverless Application Framework (SAF) and other AWS services such as Step Functions • Explore a rich variety of real-world Serverless use cases with Lambda and see how you can apply it to your environments In Detail AWS is recognized as one of the biggest market leaders for cloud computing and why not? It has evolved a lot since the time it started out by providing just basic services such as EC2 and S3 and today; they go all the way from IoT to Machine Learning, Image recognition, Chatbot Frameworks, and much more! One of those recent services that is also gaining a lot of traction is AWS Lambda! Although seemingly simple and easy to use, Lambda is a highly effective and scalable compute service that provides developers with a powerful platform to design and develop Serverless event-driven systems and applications. The book begins with a high-level introduction into the world of Serverless computing and its advantages and use cases, followed by a deep dive into AWS Lambda! You'll learn what services AWS Lambda provides to developers; how to design, write, and test Lambda functions; as well as monitor and troubleshoot them. The book is designed and accompanied with a vast variety of real-world examples, use cases, and code samples that will enable you to get started on your Serverless applications quickly. By the end of the book, you will have gained all the skills required to work with AWS Lambda services! Style and approach This step-by-step guide will help you build Serverless applications and run Serverless workloads using the AWS Lambda service. You'll be able to get started with it in a matter of minutes with easy-to-follow code snippets and examples. http://dlvr.it/R0nlzy

[ryadel]

Top Mobile App Development Trends in 2019

Over recent years, mobile app development has taken off like never before. Due to which demand for experienced mobile app development capability has increased like never before. In addition to this, businesses irrespective of their sizes are found struggling to keep up with an ever-growing backlog of projects. An untold pressure on app developers seems to be spreading across the globe to deliver ‘Best in the market’ apps. The following post emphasizes on the advent of new trends and improvement in the existing ones in the mobile realm. Technology has brought us to a stage where mobile phones being no less than a wallet can easily slide in our pockets. Impact of these smart devices can be found everywhere in all possible forms, from fitness apps to the application helping you find somebody to cuddle with in the same area. If you wish to book a cab, you have Uber, Ola at your service and if you want to order food, you have GrubHub or searching for a place to rent via Airbnb. It may quite interest you to know that the total number of smartphone users are expected to reach about 2.5 billion in 2019 from 2.1 billion in 2016. With the emergence of the internet of things (IoT), iBeacon technology, machine learning, Augmented Reality, and virtual reality and others, mobile has become a ubiquitous technology that people can’t exist without.

 #1: Accelerated Mobile Pages through AMP and EMM

After getting integrated into Google search in the year 2016, AMP listings have become very important for mobile app developers all across the globe. After this, professionals have been able to use this version of HTML in particular for better user experience and retention. Such monitoring tools and metrics are not just meant to resolve bottlenecks but also enhance overall performance to become a staple by 2019.

#2. Artificial Intelligence will change the tech

We all are well aware regarding the fact that how advancement in AI has changed the way we interact with mobile apps. Several virtual assistants like Siri, Cortana, and Google assistant can be considered while gathering information, book an appointment, optimize our productivity or organize our schedules. With advancement in Natural Language Processing and machine learning make virtual assistants more contextually aware, making their responses better and more accurate. Tech giants like Facebook, Microsoft, Google, and Amazon seem to have invested in solutions mainly based on machine learning. This definitely creates a better conversational interface.

 #3. Security will always remain a top concern

2019 is considered as a year where security becomes a major concern for everyone including businesses as well as individuals. We all know how billions of users and 3rd party SDKs and the vast number of user data are being stored by companies day in day out. This made a security a top concern for businesses as well as mobile app developers worldwide. Several encrypted messaging apps such as Telegram makes use of mobile browsers protecting users’ privacy and an industry shift towards protecting user data- all thanks to Apple’s App Transport Security (ATS) and Google’s efforts.

#4. Have you met these hyped technologies?

I am sure you must have heard about technologies like AR and VR. The futuristic technology is not only grooming and augmenting the popular advanced gaming apps but has also created a hype around the social media platforms. Pokemon GO, one of the best example that everyone knows about. In fact, social media giants like Instagram and Snapchat had started implementing this technology – user engagement has surged, and the apps have indeed become more interactive. It may also interest you to know that augmented reality is the best tool that can be used for social media campaigns through which anyone can turn any human face into a digital character.

5. Chatbots

Chatbots have become an integral part of most of the on-demand apps. These apps allow you to respond quickly in real-time to your customers. Other than this, you can also find virtual assistants the ones which does not involve human-to-human interaction. According to notable resources, chatbots will make a severe increase to 1250 million by 2025. Chatbots are well getting connected with the AR and VR technologies to write new success stories and offering the business enterprises a digital weapon to give an edge over others. It is also right to say that they have played a pivotal role in escalating the customer relationship management to a new level.

#6. Instant Apps

Within the span of these two years, instant apps have gone increasingly popular these days. These are the native mobile apps that you don’t need to download. They are user-friendly, convenient including having a smaller size. The key reasons why they are becoming more popular is that you don’t need to install them which will make your phone memory free. However, you can share it with your friends. Instant apps like New York Times Crossword, Buzzfeed, Red Bull TV, and Onefootball exempts you to download and installation that allows you to save a lot of time as well.

#7. Mobile cloud computing

2019 is here! I suppose we will be more likely to see ever-growing numbers of tools and services primarily based on cloud technology. Right from Software-as-a-Service to Infrastructure-as-a-Service, and Platform-as-a-Service, these services are expected to unfold many fronts in the upcoming years. The introduction of mobile cloud computing (MCC) solutions has led to the idea of offering rich mobile apps featuring seamless user experience irrespective of the device. More and more companies will find it easy to serve their users on mobile. However, it definitely requires more computing power that might even slow down the experience on these devices. So what can be done? Try considering mobile healthcare apps for a while. Examine how they process information to provide a remote diagnostics as part of their services.

Final Words

Last but certainly not the least, these trends are definitely going to shape how mobile websites and applications look or make us feel in future. So whether you think of creating your next chat app, photo-sharing platform or carpooling service, or any other one applicable for you, make sure to come up with one of a kind. Do you want to build a mobile application for your business? Feel free to get in touch today! Read the full article

[savetopnow]

2018-03-14 12 ECOMMERCE now

ECOMMERCE

A Better Lemonade Stand

The Ecommerce Guide to Shipping Boxes: Shipping Boxes & Mailing Envelopes 101

MOQ Meaning: What Is MOQ & What You Need to Know

Email Popups for Ecommerce: How to Collect Emails & Grow Your Sales

Scout Spotlight: Increase Conversions by Developing Strong Customer Relationships

Drop Shipping 101: The Definitive Guide to Building a Drop Shipping Business in 2018

BigCommerce

How I Turned My Father’s 90s B2B Business into California’s Fastest Growing Online Succulent Seller

Promotional Effectiveness Metrics & Email Capture Benchmarks Across 10 Ecommerce Industries [2018 Report]

Word of Mouth Marketing: How to Create a Strategy for Social Media Buzz & Skyrocket Referral Sales

How to Set Up an Ecommerce Loyalty Program to Improve Retention, Build Community and Drive 5X in Sales

How 29 Sites Built the Perfect Ecommerce Product Pages (Best Practices & Examples You Can Steal)

Bootstrapping Ecommerce

How to Create A Return Policy That Will Make Your Buyers Come Back

Budget Sales Strategy: How to Increase Your Sales on a Budget

4 Steps to Start an Online Store With a Tiny Budget

How to Find Dropshipping Products that Sell Like Crazy

The Best Business Books of All Time: 15 Books Every Entrepreneur Needs to Read

E-Commerce Times

Data Storage, Privacy and Metaphysics: SC Weighs Arguments in MS Case

The Painfully Persistent B2B Sales and Marketing Divide

The VR Experience: Challenges for a Growing Market

Amazon Courts Medicaid Recipients With Deep Prime Discount

Pega Enlists AI to Coach Sales Teams

My Wife Quit Her Job

How To Use Facebook Messenger To Sell Online – A Guide To Building Chatbots For Ecommerce

199: The Best Way To Source Private Label Products From China With Nathan Resnick Of Sourcify

198: The Software Tools I Use To Run My 7 Figure Blog With Steve Chou

197: A Better Way To Launch Products On Amazon With Scott Voelker

The Dangers Of Selling On Amazon And Horror Stories From Real Amazon Sellers

Nerd Marketing

Podcast 35: Top Nerd Travel Tools

Podcast 34: Productivity Insights After Six Months on the Road

Podcast 33: The Podcast is Back and Three Updates for 2018

The Five Simple Steps to Lose Money Over the Holidays

Black Friday Ecommerce: The Hype and the Reality

Reddit ecommerce

My site visitors are not converting to sales. Help?

Need help for building an app

Drop shipping Tariffs & laws for Egyptian cotton bed sheets.

Im thinking to dropship in amazon

Any body have experience with 3-party shipping providers?

Sleeknote

Upselling for E-Commerce: How to Increase Your Average Order Value

Use These 10 Proven Product Photography Tips to Drive More Conversions

Harry’s Marketing Revealed: What We Learned from Reverse Engineering a $400 Million Startup (Case Study)

A Definitive Guide to Lead Nurturing Emails for E-Commerce

Behavioral Marketing 101: The 3 Best Personalization Tactics You Haven’t Tried

Store Growers

How I Set Up Google Adwords Campaigns For My Online Store

How Elon Musk’s New Rocket Helped My Online Store Take Off

Building From Scratch – January 18 Update

The Confusing Thing About Google Shopping Campaign Priority

How To Do Keyword Research For Google Adwords

Traffic and Sales

5 Tips for Designing a High Converting Website Pop Up (Plus 4 Bonus Conversion Tactics)

3 No-Cost eCommerce Marketing Strategies to Increase Conversions to your New Shopify Store

The Importance of Shopper Loyalty and Loyalty Programs

How to Increase your Sales with Better Product Photos

Stick With It

[craigbrownphd-blog-blog]

Anatomy of Chatbots

Introduction Natural language conversation is one of the most challenging artificial intelligence problems, which involves language understanding, reasoning, and the utilization of common sense knowledge. Previous works in this direction mainly focus on either rule-based or learning-based methods. These types of methods often rely on manual effort in designing rules or automatic training of model with a particular learning algorithm and a small amount of data, which makes it difficult to develop an extensible open domain conversation system. Chatbots (also known as Conversation Agents or Dialog based Agents) make use of Natural Language conversation models. They are the latest trend. All big guns investing a lot on building a Chatbot to allure customer because it adds a coolness factor to any system/ website/ application of being able to solve problems in more interactive way. Companies like Microsoft, Facebook (M), Apple (Siri), Google, WeChat, and Slack are heavily investing in building their version of bots Many companies are hoping to develop bots to have natural conversations that are as similar to human ones as possible, and many are claiming to be using NLP and Deep Learning techniques to make this possible. But with all the hype around AI it’s sometimes difficult to tell fact from fiction. In this article I will try to uncover what a chatbot consists of. Chatbot Architecture Any Chatbot can consist of the following components: * Multi-channel User Interface * Communication Mediator * NLP Engine also known as Conversation model * Conversational Corpus (for training, testing and further analysis) * Interfaces to business domains Multi-channel User Interface This part of chatbot is opened to end user. User interacts with the bot from UI. The conversation can happen via multi-channels interfaces like phones, laptops, hardware (Amazon echo), kiosks, desktops etc. and it can be text based (messengers) or verbal (Siri, Alexa etc.). The user interface is responsible for providing these capabilities through which the user can interact and use the system. It is also responsible for maintaining the user session, keeping track of user activity, manage authentication/ authorization and user cart (if applicable). Communication Mediator The communication mediator of chatbot is responsible for channelizing the input coming in from User Interfaces to invoke the service that can resolve the query. In case of verbal communication, it should also be able to convert speech to text and while responding should have text to speech generation capability. This can either be done by invoking third party API (Nuance’s ASR/TTS API, Alexa API etc) or we can have pre-generated speech files (in case of rule-based models). This layer should also be responsible for making sure right model is invoked to further process the input and generate the appropriate responses. Some people consider UI and Communication as one entity but I think they are separate because they have different functions. I have also seen a few architects calling this layer as conversation state machines, knowledge as service (KaaS – a term often used by a good friend of mine @Brian Martin), or intent resolver Conversation Models Models are the brain of bot. The models help in figuring out the intent of input. In ideal scenario the model should behave as human brain that understands what a person is asking for by relating events from context it builds over the period of conversation and signals what response to be sent. The Chatobt can be modeled in various ways. Depending upon the availability of data and domain knowledge one can build a high performing bot. Here are few commonly used models Rule-based models These models are easier to build as we predefine the set of sentences (question or responses) and use some kind of heuristic to select the appropriate response based on input context. Depending upon the type of problem, we can model the heuristics to be as simple as some rules or as complex as machine learning classifiers. These systems are mostly hard-coded and they pick responses from a fixed set stored either in database or file system. It needs lot of ground work to prepare and cover all types of scenarios. Template based models These models are similar to rule based but here we just predefine the frame of the response and leave spaces for placeholders (tokens) that we can fill in based on logic we wrote. The logic can either use deep learning techniques or can be fetched from a query based system. E.g. We can have template of telling weather as – “Weather of is and temperature will go F”. In this simple example we have 3 place holders and we can easily fetch the information if the question being asked – “What is the weather in San Francisco?”. We can generalize these templates to train our model. Most of the existing chatbots make use of template based approach however the companies are constantly trying to make smarter templates so that they can get away from this hardcoding of finding too many scenarios and generate more human like responses. Generative models This is a new breed of models where we generate the responses using Machine Translation techniques. The deep neural networks using sequence-to-sequence modeling are commonly used to generate the response. Another widely used (and researched) model is Autoencoders that make use of encoder-decoder paradigm to generate responses. These systems are much smarter. Here we don’t rely on pre-defined responses and everything is done on the fly. These models are slowly becoming popular. The problem is they need a lot of data to train upon and they are still not able to generate fully grammatically correct sentences also they can’t generate large responses with high accuracy. If we have short answers they are pretty good. Conversation length based models Sometimes the need is to generate one-liners and sometimes the expectation is to generate a full paragraph of response. The bots can be modeled to generate either but in both the cases we need to handle scenarios differently. Long conversation type bots are harder to build. We can make use of deep neural network of Generators (using LSTM, GRU) capable of generating text summary for this. Short answers can follow one of the above discussed approaches. Since the response is small and pretty much follow a pattern most of the times we use rule-based or template based approach. For the answers that are highly open-ended we would need to make use of generators. Conversation type based models This class has to categories – Open ended and closed ended conversations. The model we build for these types of bots should be trained accordingly on the type of responses it need to send back. In an open ended setting the user can take the conversation anywhere. There isn’t necessarily have a well-defined goal or intention. Conversations on social media sites like Twitter and Reddit are typically open ended – they can go into all kinds of directions. The infinite number of topics and the fact that a certain amount of world knowledge is required to create reasonable responses makes this a hard problem. In a closed ended setting the space of possible inputs and outputs is somewhat limited because the system is trying to achieve a very specific goal. Technical Customer Support or Shopping Assistants are examples of closed ended problems. These systems don’t need to be able to talk about politics, they just need to fulfill their specific task as efficiently as possible. Sure, users can still take the conversation anywhere they want, but the system isn’t required to handle all these cases – and the users don’t expect it to. We can mix and match these models to get more optimum results. Having worked on generative and rules/ template based the models, I feel that generative models still have a long way to go. Even though I was able to get good number of one liner/ word answers (94% accuracy), in real world this is not sufficient. Conversational Corpus It is the bread and butter of bots. Model need to be trained on the business specific data. The model consumes the data and can become smarter. The more specific data we have the better a bot can perform. In case of building rule-based or template based bots, we can train on publically available engines like wit.ai or if we are working on generative models we need a lot of business specific data to train the custom model. Interfaces to enterprise applications The bot architecture should be pluggable with exposed interfaces that connect the bot with enterprise applications and help in catering to the purpose for which it was made. Once bot figures out what user is looking for they can programmatically invoke these interfaces to further add specifics to the responses product information in a shopping cart, inventory list or showing last n orders. The interfaces can also represent content delivery networks that can feed media and other business specific data to bot application. Conclusion In this article, I tried to cover some basic nut bolts of a chatbot. It takes a lot to build a chatbot. It needs proper planning, thought process, business domain understanding, content building, model selection, data gathering, scenarios and use case preparation and interfaces identification. The deep network technology is evolving and new methods are discovered every other day. I have built bots using most of the modeling methods I described in this article. I have strong liking for generative models because they can be made smart and they can awe you upon the type of responses they are able to generate, but they are still far away from being foolproof. I am continuing my research on building a smart bot. For developing high performing bots, still rule and template based models are extensively used. We are in pursuit of making a high performing chatbot. The bot that can resolve the issues right away with appropriate reasoning unlike the answer “42” given by “Deep Thought” in "Hitchhiker's Guide to Galaxy". I am confident that we will get there soon. https://goo.gl/rBGBdW #DataScience #Cloud