#Generative AI Best Practices

[kariniai]

Strategic Approaches to Generative AI Adoption in Enterprises

In the past twelve months, the corporate landscape has been abuzz with the potential of generative AI as a groundbreaking innovation. Despite broad recognition of its transformative power, many firms have adopted a tentative stance, cautiously navigating the implementation of this technology.

Is a cautious approach prudent, or does it inadvertently place companies at risk of lagging in a rapidly evolving technological landscape?

Recent investigations forecast the staggering benefits of generative AI, suggesting potential productivity gains in trillions of dollars per annum by 2030 if harnessed effectively.

The rewards surpass the apprehensions, provided the adoption of this technology is executed with strategic foresight. It's not about restricting generative AI but about sculpting its usage within well-defined parameters to mitigate potential challenges, including uncontrolled expenses, security breaches, compliance issues, and employee engagement.

Below, we outline ten strategic approaches for enterprises to capitalize on generative AI effectively and securely.

Adopt a Streamlined Approach to Business Case Development: Generative AI, an emerging technology, demands a departure from traditional business case development. Enterprises should prioritize rapid experimentation and learning to pinpoint practical technology applications swiftly. Discover and Explore

Action Points:

Accelerate pilot projects and proof-of-concept initiatives to cultivate knowledge and skills.

Discover and Explore and Test on repeat

Avoid:

Postponing initiatives due to the need for more absolute clarity.

Over-reliance on cumbersome business case development processes.

Initiate with Straightforward Applications: Before venturing into more complex applications, begin by unlocking value within existing business processes.

Action Points:

Concentrate on internal applications as foundational steps.

Prioritize data readiness for customized solutions.

Avoid:

Early deployment of customer-facing applications due to higher associated risks.

Use case lock where you’re working to solve a specific problem in one particular way.

Streamline Technology Evaluation: Most generative AI tools offer similar capabilities, rendering extensive evaluation unnecessary.

Action Points:

Collaborate with firms like Karini.ai for initial use cases whose platform provides immediate access to no-code tools for operationalizing Gen AI smartly.

Focus on trust and integration capabilities that open your LLMs, Models, and Data to all available options.

Avoid:

Elaborate and potentially outdated analysis of technology providers.

Vendor lock on a single platform that will cause crippling limitations.

Harness External Expertise: The scarcity of AI expertise necessitates partnerships for successful implementation and integration.

Action Points:

Assess internal expertise gaps, seek external support accordingly, and embrace a low-code/no-code platform, i.e., Karini.ai, which will keep the journey quick and safe.

Facilitate technology assimilation into the enterprise.

Avoid:

Isolated attempts at implementation.

Restrictive partnerships limit future technological choices.

Design a Flexible System Architecture: Architectures must be dynamic to accommodate evolving technologies, use cases, and regulatory landscapes.

Action Points:

Foster innovative and forward-thinking architectural design.

Anticipate and plan for future architectural adjustments.

Avoid:

Rigid architectures based on present-day technology functioning.

Over-reliance on existing processes for future technology support.

Implement Robust Security Protocols: Addressing generative AI's unique security challenges through custom policies and robust partnerships.

Action Points:

Develop tailored policies and procedures.

Partner with platforms that are active protectors of your data security.

Avoid:

Dependence on outdated security frameworks.

Technology adoption paralysis due to fear of risk.

Establish Innovative KPIs: New KPIs should reflect generative AI's unique value and impact on business operations.

Action Points:

Develop KPIs centered around long-term value creation.

Learn from both successes and failures.

Avoid:

Ignoring the learning opportunities presented by unsuccessful initiatives.

Foster Open Communication: Ensure continuous feedback and open communication channels for iterative improvement and employee engagement.

Action Points:

Integrate feedback mechanisms into all AI systems, like Karini uses in our CoPilot. 👍👎💬

Maintain transparent communication about AI's impact on the workforce.

Avoid:

Relying solely on conventional feedback methods.

Promote Comprehensive Learning and Development: Equip employees with the necessary skills and understanding to leverage AI tools effectively.

Action Points:

Provide extensive learning opportunities; Gen AI is empowering.

Align learning initiatives with broader change management strategies.

Avoid:

Limiting learning opportunities to direct users of AI tools AI needs to be democratized.

Embrace Iterative Learning: Cultivate a learning and continuous improvement culture to maximize the value derived from generative AI.

Action Points:

Prioritize learning and skill enhancement.

Engage in iterative development to refine use cases and technology applications.

Avoid:

Pursuing overly ambitious initial use cases.

Disregarding the evolving nature of AI technologies.

As enterprises stand at the cusp of this generative AI revolution, adopting a 'wait-and-see' approach may inadvertently place them at a competitive disadvantage.

The promise of generative AI far overshadows the perceived risks, demanding proactive engagement rather than cautious observation. Now is the opportune moment for enterprises to embrace generative AI, navigating its introduction with calculated measures to offset potential risks.

For further insights, explore our website or engage with our team. 

About us:

Fueled by innovation, we're making the dream of robust Generative AI systems a reality. No longer confined to specialists, Karini.ai empowers non-experts to participate actively in building/testing/deploying Generative AI applications. As the world's first GenAIOps platform, we've democratized GenAI, empowering people to bring their ideas to life – all in one evolutionary platform.

Contact us:

Jerome Mendell

(404) 891-0255

sales@karini.ai

[alwayshinny]

Hinny 💘 - The only poem Harry remembered by heart

His eyes are as green as a fresh pickled toad, 

His hair is as dark as a blackboard. 

I wish he was mine, he's really divine,

The hero who conquered the Dark Lord.

-Chamber of Secrets, page 178

[carionto]

The duality of Man, or triality? quadrality?

Alien to Human about New Human: Correct me if I'm wrong, but they appear abnormally large for your species?

H: Yea, he's a biggun alright, even without the EV suit I'd say... 7'3'', 310 pounds, bet he power lifts.

A: Umm... not to be rude, but, uhh... he seems, well... how should I put this...

H: Intimidating? Terrifying? Evil? Yea, if this station didn't have high screening standards I'd be totally pissing myself if he started walking towards me. The mohawk and eye tat totally make me believe he could snap me in two with a single glare.

A: I feel ashamed that my instincts are telling me to flee. I wish nature were easier to change.

H (shouting at NH): Hey buddy! Could you come over here for a minute please? You look awesome by the way!

A (whispering nervously): what are you doing?!?

H: Gotta overcome those fears somehow, I believe the best way is a direct confrontation.

NH approaches, somewhat slowly, looking around at all the other aliens in the station that are chatting, waiting around, or doing some work. He finally approaches A and H, and in a very deep and husky voice says: Um, hi, hello. T-thanks for the compliment, I, uh, was a little worried I would stand out too much here.

H: Oh you totally do, my friend over here is practically about to pass out from how much like a gothic viking of death metal you look.

NH: Oh no, I'm so sorry, I-I just grew up in Sweden-Delta and both my parents were huge into classic local music, so I just, uh... it's complicated. I'm sorry, I didn't mean to scare anyone.

H: Hey, relax pal, we're all good people here. Anyway, what you here to do? I'm planning on starting a bakery, still testing out what kind of flour most species here can actually stomach. My friend here is on the team working on Moon theft preventative measures.

NH: Oh, that's cool. I'm here as an exchange student with the department of applied astrophysics. If all goes well, I can finish my Bachelors degree remotely and stay here as an intern with the head researcher.

H: Oooh, that's cool. (so cool yea that you're apparently half my age but oh well guess I'm a big fat time waster like my father before me and oh god change the subject before I get depressed in front of strangers) That's a real big bag you got there, carrying some super secret science things, eh?

NH: Oh, that... uhh... guess it can't hurt to tell, security vetted it already anyway.

NH proceeds to unzip the bag and hold up a large white piece of clothing with light blue rings and accents, alongside a strange white cap with what looked like small fins, and a curious little backpack.

NH: It's uhh... um... my... Ika... musume... cosplay.... (oh gods I can't believe I said it out loud again)

After a moment of awkward silence, NH slowly puts on the backpack and presses a button on it's strap, and suddenly numerous light blue colored tentacle-like appendages sprout out from the backpack and move in line with NH's movements.

NH: I, uh..., got my engineering friend to make them articulate and interface with my contacts. I can make them do all sorts of things, like make various shapes and animals with them, though works best as a shadow theater.

H:...

NH:...

A now frozen out of confusion than fear:...

H: That's so

NH: (oh I know it's so lame, but I love that show)-

H: COOL! I don't know what a ika musume is, but those things look amazing. You said articulate? How precise can they be? I'd love to have something like that instead of my useless assistant. Poor lad can't make a piece of toast if his life depended on it...

NH: Y-you like it?

H: I LOVE those things. My daughter does cosplay too sometimes, but she makes her Dreadnought suits herself from scraps. One time the military came to our house and installed a limiter on the gauss cannon she found in a crash site, said it would otherwise start to generate small doses of radiation if used too frequently. But she replaced it with a handmade rail gun before the next convention. Do you go to those? Did you see a 7 meter tall hulking metal monstrosity with a bunch of candles all over? That was her.

NH: Oh, I think I've seen video of that, but no, not in person, I go to smaller events. I don't really like big crowds.

H: Oh yea, I get ya, you do seem a bit on the shy side now that we've been talking for a bit. Hey, no worries, like I said, we're all good people here.

NH: T-thanks, but I think I should be going now, the teacher is calling me over.

H: Oh yea, go ahead, didn't mean to take up so much of your time. Have a fun stay and I'm sure you'll ace that paper or theory? Or whatever astrophysicists do, you seem like a solid kid.

NH: Oh, uh, thanks. Good luck with your bakery. And you with stopping those weird people from stealing more moons. Bye.

H: Bye bye, come visit, don't be a stranger now, I'm set up just a short bit from the main lift on floor 14.

NH: R-right, I'll, uh, be sure to stop by soon.

A is finally able to process what they just heard and says: What was all that just now?

H: What? Just a friendly chat with what is apparently basically a kid. Man, this kid's got so much going on, while I'm almost 50 and I have an oven. Life, man, it can go in so many ways. Anyway, let's go grab a drink, I'm parched.

[digitaldeeptech]

Must-Have Programmatic SEO Tools for Superior Rankings

Understanding Programmatic SEO

What is programmatic SEO?

Programmatic SEO uses automated tools and scripts to scale SEO efforts. In contrast to traditional SEO, where huge manual efforts were taken, programmatic SEO extracts data and uses automation for content development, on-page SEO element optimization, and large-scale link building. This is especially effective on large websites with thousands of pages, like e-commerce platforms, travel sites, and news portals.

The Power of SEO Automation

The automation within SEO tends to consume less time, with large content levels needing optimization. Using programmatic tools, therefore, makes it easier to analyze vast volumes of data, identify opportunities, and even make changes within the least period of time available. This thus keeps you ahead in the competitive SEO game and helps drive more organic traffic to your site.

Top Programmatic SEO Tools

1. Screaming Frog SEO Spider

The Screaming Frog is a multipurpose tool that crawls websites to identify SEO issues. Amongst the things it does are everything, from broken links to duplication of content and missing metadata to other on-page SEO problems within your website. Screaming Frog shortens a procedure from thousands of hours of manual work to hours of automated work.

Example: It helped an e-commerce giant fix over 10,000 broken links and increase their organic traffic by as much as 20%.

2. Ahrefs

Ahrefs is an all-in-one SEO tool that helps you understand your website performance, backlinks, and keyword research. The site audit shows technical SEO issues, whereas its keyword research and content explorer tools help one locate new content opportunities.

Example: A travel blog that used Ahrefs for sniffing out high-potential keywords and updating its existing content for those keywords grew search visibility by 30%.

3. SEMrush

SEMrush is the next well-known, full-featured SEO tool with a lot of features related to keyword research, site audit, backlink analysis, and competitor analysis. Its position tracking and content optimization tools are very helpful in programmatic SEO.

Example: A news portal leveraged SEMrush to analyze competitor strategies, thus improving their content and hoisting themselves to the first page of rankings significantly.

4. Google Data Studio

Google Data Studio allows users to build interactive dashboards from a professional and visualized perspective regarding SEO data. It is possible to integrate data from different sources like Google Analytics, Google Search Console, and third-party tools while tracking SEO performance in real-time.

Example: Google Data Studio helped a retailer stay up-to-date on all of their SEO KPIs to drive data-driven decisions that led to a 25% organic traffic improvement.

5. Python

Python, in general, is a very powerful programming language with the ability to program almost all SEO work. You can write a script in Python to scrape data, analyze huge datasets, automate content optimization, and much more.

Example: A marketing agency used Python for thousands of product meta-description automations. This saved the manual time of resources and improved search rank.

The How for Programmatic SEO

Step 1: In-Depth Site Analysis

Before diving into programmatic SEO, one has to conduct a full site audit. Such technical SEO issues, together with on-page optimization gaps and opportunities to earn backlinks, can be found with tools like Screaming Frog, Ahrefs, and SEMrush.

Step 2: Identify High-Impact Opportunities

Use the data collected to figure out the biggest bang-for-buck opportunities. Look at those pages with the potential for quite a high volume of traffic, but which are underperforming regarding the keywords focused on and content gaps that can be filled with new or updated content.

Step 3: Content Automation

This is one of the most vital parts of programmatic SEO. Scripts and tools such as the ones programmed in Python for the generation of content come quite in handy for producing significant, plentiful, and high-quality content in a short amount of time. Ensure no duplication of content, relevance, and optimization for all your target keywords.

Example: An e-commerce website generated unique product descriptions for thousands of its products with a Python script, gaining 15% more organic traffic.

Step 4: Optimize on-page elements

Tools like Screaming Frog and Ahrefs can also be leveraged to find loopholes for optimizing the on-page SEO elements. This includes meta titles, meta descriptions, headings, or even adding alt text for images. Make these changes in as effective a manner as possible.

Step 5: Build High-Quality Backlinks

Link building is one of the most vital components of SEO. Tools to be used in this regard include Ahrefs and SEMrush, which help identify opportunities for backlinks and automate outreach campaigns. Begin to acquire high-quality links from authoritative websites.

Example: A SaaS company automated its link-building outreach using SEMrush, landed some wonderful backlinks from industry-leading blogs, and considerably improved its domain authority. ### Step 6: Monitor and Analyze Performance

Regularly track your SEO performance on Google Data Studio. Analyze your data concerning your programmatic efforts and make data-driven decisions on the refinement of your strategy.

See Programmatic SEO in Action

50% Win in Organic Traffic for an E-Commerce Site

Remarkably, an e-commerce electronics website was undergoing an exercise in setting up programmatic SEO for its product pages with Python scripting to enable unique meta descriptions while fixing technical issues with the help of Screaming Frog. Within just six months, the experience had already driven a 50% rise in organic traffic.

A Travel Blog Boosts Search Visibility by 40%

Ahrefs and SEMrush were used to recognize high-potential keywords and optimize the content on their travel blog. By automating updates in content and link-building activities, it was able to set itself up to achieve 40% increased search visibility and more organic visitors.

User Engagement Improvement on a News Portal

A news portal had the option to use Google Data Studio to make some real-time dashboards to monitor their performance in SEO. Backed by insights from real-time dashboards, this helped them optimize the content strategy, leading to increased user engagement and organic traffic.

Challenges and Solutions in Programmatic SEO

Ensuring Content Quality

Quality may take a hit in the automated process of creating content. Therefore, ensure that your automated scripts can produce unique, high-quality, and relevant content. Make sure to review and fine-tune the content generation process periodically.

Handling Huge Amounts of Data

Dealing with huge amounts of data can become overwhelming. Use data visualization tools such as Google Data Studio to create dashboards that are interactive, easy to make sense of, and result in effective decision-making.

Keeping Current With Algorithm Changes

Search engine algorithms are always in a state of flux. Keep current on all the recent updates and calibrate your programmatic SEO strategies accordingly. Get ahead of the learning curve by following industry blogs, attending webinars, and taking part in SEO forums.

Future of Programmatic SEO

The future of programmatic SEO seems promising, as developing sectors in artificial intelligence and machine learning are taking this space to new heights. Developing AI-driven tools would allow much more sophisticated automation of tasks, thus making things easier and faster for marketers to optimize sites as well.

There are already AI-driven content creation tools that can make the content to be written highly relevant and engaging at scale, multiplying the potential of programmatic SEO.

Conclusion

Programmatic SEO is the next step for any digital marketer willing to scale up efforts in the competitive online landscape. The right tools and techniques put you in a position to automate key SEO tasks, thus optimizing your website for more organic traffic. The same goals can be reached more effectively and efficiently if one applies programmatic SEO to an e-commerce site, a travel blog, or even a news portal.

[pinkautist]

had to unfollow and artist i enjoyed bc they talked about how they use ai and took the stance of, "people need to stop attacking me for it bc it's been really helpful to me as a disabled artist 🥺 we should be standing together as artists not trying to divide ourselves 🥺" you know what else ai does? YOU KNOW WHAT ELSE AI DOES??? IT USES ENOUGH ENERGY TO KILL OUR PLANET MUCH FASTER THAN ANYTHING ELSE WE CAN DO ENERGY-WISE.

i am a disabled artist too. my cognitive and mental disabilities that make burnout a substantial roadblock aside, i can have frequent and extreme pain in my wrist (sometimes out of nowhere, sometimes as a result of drawing) that makes drawing an extremely painful and slow process for me (this is due to hypermobility probably). sometimes i draw in spite of the pain, and sometimes it's enough to have me taking long breaks. as a result of the combination of disabilities i have, i am an extremely slow artist. sometimes i struggle with having intense motivation to create, but no actual inspiration because of things going on in my head. and it's a struggle because i desperately want to create, but nothing is coming to me, which makes me feel really bad. and in spite of ALL of this, i still will not use fucking ai to "soothe the uninspired motivation" or to "create in spite of my pain". i will not hand my humanity over to ai because of such trivial reasons. ai could never do what i can, because creating is about the process, not the end result. i feel accomplished in a way that using ai could never provide when i see a piece coming together, something that i've created from nothing. ai will never provide that feeling.

[cubicdigital]

How AI Content Affects Google Search Rankings in 2025

Discover how AI-generated content influences Google's search rankings in 2025. This in-depth article explores Google's stance on AI content, the role of E-E-A-T guidelines, and best practices for maintaining or improving SEO. Learn how to balance automation with human insight to stay competitive in search results while aligning with Google's quality standards. Stay informed and future-proof your content strategy with expert tips and actionable insights.

[kotori-mochi]

Can't afford art school?

After seeing post like this 👇

And this gem 👇

As well as countless of others from the AI generator community. Just talking about how "inaccessible art" is, I decided why not show how wrong these guys are while also helping anyone who actually wants to learn.

Here is the first one ART TEACHERS! There are plenty online and in places like youtube.

📺Here is my list:

Proko (Free, mostly teaches anatomy and how to draw people. But does have art talks and teaches the basics.)

Marc Brunet (Free but he does have other classes for a cheap price. Use to work for Blizzard and teaches you everything)

Aaron Rutten (free, tips about art, talks about art programs and the best products for digital art)

BoroCG (free, teaches a verity of art mediums from 3D modeling to digital painting. As well as some tips that can be used across styles)

Jesse J. Jones (free, talks about animating)

Jesus Conde (free, teaches digital painting and has classes in Spanish)

Mohammed Agbadi (free, he gives some advice in some videos and talks about art)

Ross Draws (free, he does have other classes for a good price. Mostly teaching character designs and simple backgrounds.)

SamDoesArts (free, gives good advice and critiques)

Drawfee Show (free, they do give some good advice and great inspiration)

The Art of Aaron Blaise ( useful tips for digital art and animation. Was an animator for Disney. Mostly nature art)

Bobby Chiu ( useful tips and interviews with artist who are in the industry or making a living as artist)

Sinix Design (has some tips on drawing people)

Winged canvas (art school for free on a verity of mediums)

Bob Ross (just a good time, learn how to paint, as well as how too relax when doing art. "there are no mistakes only happy accidents", this channel also provides tips from another artist)

Scott Christian Sava (Inspiration and provides tips and advice)

Pikat (art advice and critiques)

Drawbox (a suggested cheap online art school, made of a community of artist)

Skillshare (A cheap learning site that has art classes ranging from traditional to digital. As well as Animation and tutorials on art programs. All under one price, in the USA it's around $34 a month)

Human anatomy for artist (not a video or teacher but the site is full of awesome refs to practice and get better at anatomy)

Second part BOOKS, I have collected some books that have helped me and might help others.

📚Here is my list:

The "how to draw manga" series produced by Graphic-sha. These are for manga artist but they give great advice and information.

"Creating characters with personality" by Tom Bancroft. A great book that can help not just people who draw cartoons but also realistic ones. As it helps you with facial ques and how to make a character interesting.

"Albinus on anatomy" by Robert Beverly Hale and Terence Coyle. Great book to help someone learn basic anatomy.

"Artistic Anatomy" by Dr. Paul Richer and Robert Beverly Hale. A good book if you want to go further in-depth with anatomy.

"Directing the story" by Francis Glebas. A good book if you want to Story board or make comics.

"Animal Anatomy for Artists" by Eliot Goldfinger. A good book for if you want to draw animals or creatures.

"Constructive Anatomy: with almost 500 illustrations" by George B. Bridgman. A great book to help you block out shadows in your figures and see them in a more 3 diamantine way.

"Dynamic Anatomy: Revised and expand" by Burne Hogarth. A book that shows how to block out shapes and easily understand what you are looking out. When it comes to human subjects.

"An Atlas of animal anatomy for artist" by W. Ellenberger and H. Dittrich and H. Baum. This is another good one for people who want to draw animals or creatures.

Etherington Brothers, they make books and have a free blog with art tips.

📝As for Supplies, I recommend starting out cheap, buying Pencils and art paper at dollar tree or 5 below. If you want to go fancy Michaels is always a good place for traditional supplies. They also get in some good sales and discounts. For digital art, I recommend not starting with a screen art drawing tablet as they are usually more expensive.

For the Best art Tablet I recommend either Xp-pen, Bamboo or Huion. Some can range from about 40$ to the thousands.

💻As for art programs here is a list of Free to pay.

Clip Studio paint ( you can choose to pay once or sub and get updates. Galaxy, Windows, macOS, iPad, iPhone, Android, or Chromebook device. )

Procreate ( pay once for $9.99 usd, IPAD & IPHONE ONLY)

Blender (for 3D modules/sculpting, animation and more. Free)

PaintTool SAI (pay but has a 31 day free trail)

Krita (Free)

mypaint (free)

FireAlpaca (free)

Aseprite ($19.99 usd but has a free trail, for pixel art Windows & macOS)

Drawpile (free and for if you want to draw with others)

IbisPaint (free, phone app ONLY)

Medibang (free, IPAD, Android and PC)

NOTE: Some of these can work on almost any computer like Clip and Sai but others will require a bit stronger computer like Blender. Please check their sites for if your computer is compatible.

So do with this information as you will but as you can tell there are ways to learn how to become an artist, without breaking the bank. The only thing that might be stopping YOU from using any of these things, is YOU.

I have made time to learn to draw and many artist have too. Either in-between working two jobs or taking care of your family and a job or regular school and chores. YOU just have to take the time or use some time management, it really doesn't take long to practice for like an hour or less. YOU also don't have to do it every day, just once or three times a week is fine.

Hope this was helpful and have a great day.

"also apologies for any spelling or grammar errors, I have Dyslexia and it makes my brain go XP when it comes to speech or writing"

[mehmetyildizmelbourne-blog]

Is SEO Dead or Simply Evolving?

A Strategic Perspective on AI, Content Search, and the Future of Content Marketing Information Management and Knowledge Dissemination While it is difficult to provide a precise count, a simple Google search for “SEO” yields millions of results, indicating the subject’s extensive coverage. This abundance indicates the importance of SEO in digital marketing and the ongoing interest in optimizing…

[appiantips]

Appian Code Optimization: Essential Techniques for Clean and Efficient Development

Writing clean code in Appian, as in any other programming environment, involves following best practices that make your code more readable, maintainable, and efficient. Appian is a low-code automation platform, so while some practices are general to all coding, others are specific to Appian’s environment and tools. Here are some key practices for writing clean code in Appian: 1. Use Consistent…

[daivignanmarketingservices]

Next-Level SEO: Mastering Generative Engine Optimization Techniques

Next-Level SEO: Mastering Generative Engine Optimization Techniques

Generative Engine Optimization (GEO) ExplainedDefinition:Step-by-Step Process for Generative Engine Optimization (GEO)1. Understand Your Audience and Goals2. Keyword Research and Integration3. Content Generation with AI Tools4. Content Quality Assurance5. Metadata Optimization6. Incorporate Multimedia Elements7. User Experience (UX) Optimization8. Content Relevance and Engagement9. Natural…

[makingqueerhistory]

":')))))))) you realise that gen AI is available to everyone though right??? Queer creators can use it just as much as anyone else??? I just don't understand this post... It really feels like a cheap way to get on the 'AI Bad's bandwagon, and coming from such a thoughtful and insightful creator that's incredibly disappointing... It's okay to not comment on subjects you're not an expert in y'know...?"

Y'all know the drill, I am replying to this publicly but that is not an invitation to send any negative messages to the person I am replying to.

Anyways, let me start by saying that the original context of the post you're replying to is discussing an event where a queer org used generative AI to steal an interview with Keri Hulme. So let's start there. To be clear I don't even know if the original interviewer was queer so let's put the identities of stealer and stolen from to the side. I want to explain the harm done in this example specifically and I hope this is illustrative of what harm generative AI can (and does) do.

The original place I saw generative AI was a queer org that explicitly says they are using generative AI "for good", and as a way to bring more queer history to light. So let's take them at their word, and assume they are not out to cause harm. This is the best example of generative AI that I can imagine, so I hope that makes it clear that I am not coming at this issue from bad faith in any way.

Here is the harm they are causing:

Decontextualizing and rephrasing an interview: I am not going to pretend that I am an expert in academic best practices, but I do believe one thing, if a person is speaking on their own identity and lived experience, it is always much better to directly quote than it is to rephrase. As I read this source, I initially didn't know that it was AI, and I was already upset. An interview that is widely available on the internet with no pay wall, was poorly sourced and made more vague than it was in the initial text. By creating one degree of seperation between the original words of A WRITER (whose literal job was largely based in choosing the right words to describe experiences they had) harm is already done. It makes vague what was once clear, and removes Keri Hulme's voice from her own narrative.

The original interviewer is not paid, or given proper recognition: I get it, sometimes just copy pasting an interview doesn't feel transformative enough, but something that one would learn if they worked in the queer history field and weren't a literal robot rehashing what has already been said, is that not everything needs to be transformed. In those cases, we give credit to the person who said the original words (in this case Keri Hulme), and the interviewer who facillitated the conversation (in this case Shelley Bridgeman). This case (again a best case scenario), takes the attention and byline away from the original interviewer and gives it to an AI.

The original publisher of this story is deinsentivised from paying interviewers in the future: The original publisher of this interview has ads on their website. As a person who also has ads on their website, taking an article like this and rephrasing it for no good reason (the orginal word count was not prohibitive and the rephrasing did not make it more readable), takes money from the publisher. It's pennies, but it's also removing numbers could have been used to justify further interviews with asexual people and archiving of asexual stories. The org that stole from this publication does not interview people themselves so the money and numbers that could have gone to continue to preserve asexual stories goes to stealing them instead.

These are just the active harms that I saw in this specific case. As you said, I am not an expert in generative AI, and will not be speaking as if I am. But I will say that asking me not to speak out on active harm that is being caused in queer history spaces, is disrespectful to my many years in this field.

To illustrate this even clearer: if you were a patron, you would know I recently took down an old article. I have been rereading and editing our backlist of articles, and I found one that no longer fit my standards of sourcing. My standards had recently raised due to a video made by HBomberguy about someone in the queer history space who was stealing from other creators. I watched this video not as a work project, but because I watch most of HBomberguys videos, and this one made me think more critically about sourcing. An AI can't do that. All an AI has is what has been inputted, and it is right now impossible to input every available peice of information about ethics into an AI and get a coherent ethical basis on which it will function.

It is a distinctly human trait to absorb information and change in that way. AI can rephrase information that already exists, steal it, recontextualize it even, but it cannot create something altogether new.

Do I believe that there one day might be an ethical use for Generative AI? Maybe. Do I believe that coming into a queer history space, stealing the words of a Maori asexual author, rephrasing them, and giving the original interviewer and publication no form of compensation for their work, is accomplishing that? No.

On a more personal note: I am coming at this issue with a bias. As a queer history creator, I do not want AI in my space, because it is literally damaging to my financial prospects. It has been like pulling teeth to try and get patrons in the current state of the global economy. I don't blame anyone from that, but I feel very disrespected that I am being asked to compete with a machine now. Not only that, but I am being asked to shut up and be fine with it? No, absolutely not. I cannot and will not stay quiet as space that I have fought tooth and nail to create in mainstream discussions is taken and given to AI.

AI was not supporting me when I was sent gore to try and scare me off of discussing queer history. A person did that. AI was not there to tell me I had written too many sad stories, and I needed some happy endings to remind myself of the good in the world. A person did that. AI was not there when I was being harrassed for supporting and including asexual stories on my website. A person did that.

And after all that, I am being asked to lie down and take it when my ability to pay the people who supported me in those ways, is being threatened. Nope. Not going to happen.

An AI doesn't have to make rent. An AI doesn't understand what it feels like to have to stop holding their wife's hand in public. An AI didn't get calls from people needing comfort in reaction to the election. Pay me for my work, and get this AI nonsense out of my face.

[hms-software-timecontrol]

Where does TimeControl stand in the use of AI?

Where does TimeControl stand in the use of AI? Further than you might think.

HMS started using AI techniques in TimeControl back in 1999.  Surprised?  Artificial Intelligence has been around awhile and you probably wouldn’t have noticed how we leveraged it back in 99.  At that time we created a communications protocol called “HMI” for TimeControl to transmit enormous volumes of TimeControl data through the Internet.  It had the capability of re-routing traffic based on…

[communistkenobi]

(taken from a post about AI)

speaking as someone who has had to grade virtually every kind of undergraduate assignment you can think of for the past six years (essays, labs, multiple choice tests, oral presentations, class participation, quizzes, field work assignments, etc), it is wild how out-of-touch-with-reality people’s perceptions of university grading schemes are. they are a mass standardised measurement used to prove the legitimacy of your degree, not how much you’ve learned. Those things aren’t completely unrelated to one another of course, but they are very different targets to meet. It is standard practice for professors to have a very clear idea of what the grade distribution for their classes are before each semester begins, and tenure-track assessments (at least some of the ones I’ve seen) are partially judged on a professors classes’ grade distributions - handing out too many A’s is considered a bad thing because it inflates student GPAs relative to other departments, faculties, and universities, and makes classes “too easy,” ie, reduces the legitimate of the degree they earn. I have been instructed many times by professors to grade easier or harder throughout the term to meet those target averages, because those targets are the expected distribution of grades in a standardised educational setting. It is standard practice for teaching assistants to report their grade averages to one another to make sure grade distributions are consistent. there’s a reason profs sometimes curve grades if the class tanks an assignment or test, and it’s generally not because they’re being nice!

this is why AI and chatgpt so quickly expanded into academia - it’s not because this new generation is the laziest, stupidest, most illiterate batch of teenagers the world has ever seen (what an original observation you’ve made there!), it’s because education has a mass standard data format that is very easily replicable by programs trained on, yanno, large volumes of data. And sure the essays generated by chatgpt are vacuous, uncompelling, and full of factual errors, but again, speaking as someone who has graded thousands of essays written by undergrads, that’s not exactly a new phenomenon lol

I think if you want to be productively angry at ChatGPT/AI usage in academia (I saw a recent post complaining that people were using it to write emails of all things, as if emails are some sacred form of communication), your anger needs to be directed at how easily automated many undergraduate assignments are. Or maybe your professors calculating in advance that the class average will be 72% is the single best way to run a university! Who knows. But part of the emotional stakes in this that I think are hard for people to admit to, much less let go of, is that AI reveals how rote, meaningless, and silly a lot of university education is - you are not a special little genius who is better than everyone else for having a Bachelor’s degree, you have succeeded in moving through standardised post-secondary education. This is part of the reason why disabled people are systematically barred from education, because disability accommodations require a break from this standardised format, and that means disabled people are framed as lazy cheaters who “get more time and help than everyone else.” If an AI can spit out a C+ undergraduate essay, that of course threatens your sense of superiority, and we can’t have that, can we?

[afeelgoodblog]

The Best News of Last Month - August 2024

1.Negative Power Prices Hit Europe as Renewable Energy Floods the Grid

European power markets are experiencing a notable shift as renewable energy sources, particularly wind and solar, become a larger part of the energy mix. On Wednesday, power prices in several European markets, including Germany, dipped below zero due to a surge in green electricity production.

2. Taiwan introduces ban on performances by captive wild animals

Live performances by wild animals held in captivity, including performances by dolphins, tigers, and other non-domesticated mammals, will no longer be permitted in Taiwan under new Ministry of Agriculture (MOA) regulations.

3. FTC bans fake online reviews, inflated social media influence; rule takes effect in October

The FTC voted unanimously to ban marketers from using fake reviews, such as those generated with AI technology, and other misleading advertising practices.

The ban also forbids marketers from exaggerating their own influence by, for example, paying for bots to inflate their follower count.

4. Chinese drones will fly trash out of Everest slopes

Come autumn, Nepal will deploy heavy lifter drones to transport garbage from the 6,812-metre tall Ama Dablam, south of Everest. This will be the first commercial work an unmanned aerial vehicle does in Nepal’s high-altitude zone.

The heavy lifter from China’s biggest drone maker, Da Jiang Innovations (DJI), will take on tasks traditionally handled by Sherpas. Officials believe it will help reduce casualties on Everest.

5. Swiss scientists have found a way to use the whole cocoa fruit to make chocolate and not just taking beans and discarding the rest.

Kim Mishra (L) and Anian Schreiber (R) cooperated on the new chocolate making process

Food scientists in Switzerland have come up with a way to make chocolate using the entire cocoa fruit rather than just the beans - and without using sugar.

The chocolate, developed at Zurich’s prestigious Federal Institute of Technology by scientist Kim Mishra and his team includes the cocoa fruit pulp, the juice, and the husk, or endocarp.

6. Six-year-old boy found in Vietnam forest after five days

A six-year-old boy who was missing for five days has been found deep in a forest in Vietnam. Dang Tien Lam, who lives in the northwestern Yen Bai province, was playing in a stream with his nine siblings on 17 August when he wandered into the hills and got lost, local reports said.

He was found on Wednesday by local farmers who heard a child's cry while they were clearing a cinnamon field close to the forest.

7. Lego plans to make half the plastic in bricks from renewable materials by 2026

Lego plans to make half the plastic in its bricks from renewable or recycled material rather than fossil fuels by 2026, in its latest effort to ensure its toys are more environmentally friendly.

The Danish company last year ditched efforts to make bricks entirely from recycled bottles because of cost and production issues. At the moment, 22% of the material in its colourful bricks is not made from fossil fuels.

----------

That's it for this month :)

This newsletter will always be free. If you liked this post you can support me with a small kofi donation here:

Buy me a coffee ❤️

Also don’t forget to share this post with your friends.

[subjectsix]

KIP'S BIG POST OF THINGS TO MAKE THE INTERNET & TECHNOLOGY SUCK A LITTLE LESS

Post last updated November 23, 2024. Will continue to update!

Here are my favorite things to use to navigate technology my own way:

A refurbished iPod loaded with Rockbox OS (Rockbox is free, iPods range in price. I linked the site I got mine from. Note that iPods get finicky about syncing and the kind of cord it has— it may still charge but might not recognize the device to sync. Getting an original Apple cord sometimes helps). Rockbox has ports for other MP3 players as well.

This Windows debloater program (there are viable alternatives out there, this one works for me). It has a powershell script that give you a little UI and buttons to press, which I appreciate, as I'm still a bit shy with tech.

Firefox with the following extensions: - Consent-O-Matic (set your responses to ALL privacy/cookie pop-ups in the extension, and it will answer all pop-ups for you. I can see reasons to not use it, but I appreciate it) - Facebook Container ("contains" Meta on Facebook and Instagram pages to keep it from tracking you or getting third party cookies, since Meta is fairly egregious about it) - Redirect Amp to HTML (AMP is designed for mobile phones, this forces pages to go to their HTML version) - A WebP/AVIF image converter - uBlock Origin and uBlacklist, with the AI blacklist loaded in to kill any generative AI results from appearing in search engines or anywhere.

Handbrake for ripping DVDs— I haven’t used this in awhile as I haven’t been making video edits. I used this back when I had a Mac OS

VLC Media Player (ol’ reliable)

Unsplash & Pexels for free-to-use images

A password manager (these often are paid. I use Dashlane. There are many options, feel free to search around and ask for recs!). There is a lot that goes into cybersecurity— find the option you feel is best for you.

Things I suggest:

Understanding Royalty Free and the Creative Commons licenses

Familiarity with boolean operators for searching

Investing in a backup drive and external drive

A few good USBs, including one that has a backup of your OS on it

Adapter cables

Avoiding Fandom “wikias” (as in the brand “Fandom”) and supporting other, fan-run or supported wikis. Consider contributing if its something you find yourself passionate or joyful about.

Finding Forums for the things you like, or creating your own*

Create an email specifically for ads/shopping— use it to receive all promotional emails to keep your inbox clean. Upkeep it.

Stop putting so much of your personal information online— be willing to separate your personal online identity from your “online identity”. You don’t owe people your name, location, pronouns, diagnoses, or any of that. It’s your choice, but be discerning in what you give and why. I recommend avoiding providing your phone number to sites as much as possible.

Be intentional

Ask questions

Talk to people

Remember that you can lurk all you want

Things that are fun to check out:

BBSes-- here's a portal to access them.

Neocities

*Forums-- find some to join, or maybe host your own? The system I was most familiar with was vbulletin.

MMM.page

Things that have worked well for me but might work for you, YMMV:

Limit your app usage time on your smartphone if you’re prone to going back to them— this is a tangible way to “practice mindfulness”, a term I find frustratingly vague ansjdbdj

Things I’m looking into:

The “Pi Hole”— a raspberry pi set up to block all ads on a specific internet connection

VPNs-- this is one that was recommended to me.

How to use computers (I mean it): Resources on how to understand your machine and what you’re doing, even if your skill and knowledge level is currently 0:

This section I'll come back an add to. I know that messing with computers can be intimidating, especially if you feel out of your depth. HTML and regedits and especially things like dualbooting or linux feel impossible. So I want to put things here that explain exactly how the internet and your computer functions, and how you can learn and work with that. Yippee!

[mostlysignssomeportents]

Even if you think AI search could be good, it won’t be good

TONIGHT (May 15), I'm in NORTH HOLLYWOOD for a screening of STEPHANIE KELTON'S FINDING THE MONEY; FRIDAY (May 17), I'm at the INTERNET ARCHIVE in SAN FRANCISCO to keynote the 10th anniversary of the AUTHORS ALLIANCE.

The big news in search this week is that Google is continuing its transition to "AI search" – instead of typing in search terms and getting links to websites, you'll ask Google a question and an AI will compose an answer based on things it finds on the web:

https://blog.google/products/search/generative-ai-google-search-may-2024/

Google bills this as "let Google do the googling for you." Rather than searching the web yourself, you'll delegate this task to Google. Hidden in this pitch is a tacit admission that Google is no longer a convenient or reliable way to retrieve information, drowning as it is in AI-generated spam, poorly labeled ads, and SEO garbage:

https://pluralistic.net/2024/05/03/keyword-swarming/#site-reputation-abuse

Googling used to be easy: type in a query, get back a screen of highly relevant results. Today, clicking the top links will take you to sites that paid for placement at the top of the screen (rather than the sites that best match your query). Clicking further down will get you scams, AI slop, or bulk-produced SEO nonsense.

AI-powered search promises to fix this, not by making Google search results better, but by having a bot sort through the search results and discard the nonsense that Google will continue to serve up, and summarize the high quality results.

Now, there are plenty of obvious objections to this plan. For starters, why wouldn't Google just make its search results better? Rather than building a LLM for the sole purpose of sorting through the garbage Google is either paid or tricked into serving up, why not just stop serving up garbage? We know that's possible, because other search engines serve really good results by paying for access to Google's back-end and then filtering the results:

https://pluralistic.net/2024/04/04/teach-me-how-to-shruggie/#kagi

Another obvious objection: why would anyone write the web if the only purpose for doing so is to feed a bot that will summarize what you've written without sending anyone to your webpage? Whether you're a commercial publisher hoping to make money from advertising or subscriptions, or – like me – an open access publisher hoping to change people's minds, why would you invite Google to summarize your work without ever showing it to internet users? Nevermind how unfair that is, think about how implausible it is: if this is the way Google will work in the future, why wouldn't every publisher just block Google's crawler?

A third obvious objection: AI is bad. Not morally bad (though maybe morally bad, too!), but technically bad. It "hallucinates" nonsense answers, including dangerous nonsense. It's a supremely confident liar that can get you killed:

https://www.theguardian.com/technology/2023/sep/01/mushroom-pickers-urged-to-avoid-foraging-books-on-amazon-that-appear-to-be-written-by-ai

The promises of AI are grossly oversold, including the promises Google makes, like its claim that its AI had discovered millions of useful new materials. In reality, the number of useful new materials Deepmind had discovered was zero:

https://pluralistic.net/2024/04/23/maximal-plausibility/#reverse-centaurs

This is true of all of AI's most impressive demos. Often, "AI" turns out to be low-waged human workers in a distant call-center pretending to be robots:

https://pluralistic.net/2024/01/31/neural-interface-beta-tester/#tailfins

Sometimes, the AI robot dancing on stage turns out to literally be just a person in a robot suit pretending to be a robot:

https://pluralistic.net/2024/01/29/pay-no-attention/#to-the-little-man-behind-the-curtain

The AI video demos that represent "an existential threat to Hollywood filmmaking" turn out to be so cumbersome as to be practically useless (and vastly inferior to existing production techniques):

https://www.wheresyoured.at/expectations-versus-reality/

But let's take Google at its word. Let's stipulate that:

a) It can't fix search, only add a slop-filtering AI layer on top of it; and

b) The rest of the world will continue to let Google index its pages even if they derive no benefit from doing so; and

c) Google will shortly fix its AI, and all the lies about AI capabilities will be revealed to be premature truths that are finally realized.

AI search is still a bad idea. Because beyond all the obvious reasons that AI search is a terrible idea, there's a subtle – and incurable – defect in this plan: AI search – even excellent AI search – makes it far too easy for Google to cheat us, and Google can't stop cheating us.

Remember: enshittification isn't the result of worse people running tech companies today than in the years when tech services were good and useful. Rather, enshittification is rooted in the collapse of constraints that used to prevent those same people from making their services worse in service to increasing their profit margins:

https://pluralistic.net/2024/03/26/glitchbread/#electronic-shelf-tags

These companies always had the capacity to siphon value away from business customers (like publishers) and end-users (like searchers). That comes with the territory: digital businesses can alter their "business logic" from instant to instant, and for each user, allowing them to change payouts, prices and ranking. I call this "twiddling": turning the knobs on the system's back-end to make sure the house always wins:

https://pluralistic.net/2023/02/19/twiddler/

What changed wasn't the character of the leaders of these businesses, nor their capacity to cheat us. What changed was the consequences for cheating. When the tech companies merged to monopoly, they ceased to fear losing your business to a competitor.

Google's 90% search market share was attained by bribing everyone who operates a service or platform where you might encounter a search box to connect that box to Google. Spending tens of billions of dollars every year to make sure no one ever encounters a non-Google search is a cheaper way to retain your business than making sure Google is the very best search engine:

https://pluralistic.net/2024/02/21/im-feeling-unlucky/#not-up-to-the-task

Competition was once a threat to Google; for years, its mantra was "competition is a click away." Today, competition is all but nonexistent.

Then the surveillance business consolidated into a small number of firms. Two companies dominate the commercial surveillance industry: Google and Meta, and they collude to rig the market:

https://en.wikipedia.org/wiki/Jedi_Blue

That consolidation inevitably leads to regulatory capture: shorn of competitive pressure, the companies that dominate the sector can converge on a single message to policymakers and use their monopoly profits to turn that message into policy:

https://pluralistic.net/2022/06/05/regulatory-capture/

This is why Google doesn't have to worry about privacy laws. They've successfully prevented the passage of a US federal consumer privacy law. The last time the US passed a federal consumer privacy law was in 1988. It's a law that bans video store clerks from telling the newspapers which VHS cassettes you rented:

https://en.wikipedia.org/wiki/Video_Privacy_Protection_Act

In Europe, Google's vast profits lets it fly an Irish flag of convenience, thus taking advantage of Ireland's tolerance for tax evasion and violations of European privacy law:

https://pluralistic.net/2023/05/15/finnegans-snooze/#dirty-old-town

Google doesn't fear competition, it doesn't fear regulation, and it also doesn't fear rival technologies. Google and its fellow Big Tech cartel members have expanded IP law to allow it to prevent third parties from reverse-engineer, hacking, or scraping its services. Google doesn't have to worry about ad-blocking, tracker blocking, or scrapers that filter out Google's lucrative, low-quality results:

https://locusmag.com/2020/09/cory-doctorow-ip/

Google doesn't fear competition, it doesn't fear regulation, it doesn't fear rival technology and it doesn't fear its workers. Google's workforce once enjoyed enormous sway over the company's direction, thanks to their scarcity and market power. But Google has outgrown its dependence on its workers, and lays them off in vast numbers, even as it increases its profits and pisses away tens of billions on stock buybacks:

https://pluralistic.net/2023/11/25/moral-injury/#enshittification

Google is fearless. It doesn't fear losing your business, or being punished by regulators, or being mired in guerrilla warfare with rival engineers. It certainly doesn't fear its workers.

Making search worse is good for Google. Reducing search quality increases the number of queries, and thus ads, that each user must make to find their answers:

https://pluralistic.net/2024/04/24/naming-names/#prabhakar-raghavan

If Google can make things worse for searchers without losing their business, it can make more money for itself. Without the discipline of markets, regulators, tech or workers, it has no impediment to transferring value from searchers and publishers to itself.

Which brings me back to AI search. When Google substitutes its own summaries for links to pages, it creates innumerable opportunities to charge publishers for preferential placement in those summaries.

This is true of any algorithmic feed: while such feeds are important – even vital – for making sense of huge amounts of information, they can also be used to play a high-speed shell-game that makes suckers out of the rest of us:

https://pluralistic.net/2024/05/11/for-you/#the-algorithm-tm

When you trust someone to summarize the truth for you, you become terribly vulnerable to their self-serving lies. In an ideal world, these intermediaries would be "fiduciaries," with a solemn (and legally binding) duty to put your interests ahead of their own:

https://pluralistic.net/2024/05/07/treacherous-computing/#rewilding-the-internet

But Google is clear that its first duty is to its shareholders: not to publishers, not to searchers, not to "partners" or employees.

AI search makes cheating so easy, and Google cheats so much. Indeed, the defects in AI give Google a readymade excuse for any apparent self-dealing: "we didn't tell you a lie because someone paid us to (for example, to recommend a product, or a hotel room, or a political point of view). Sure, they did pay us, but that was just an AI 'hallucination.'"

The existence of well-known AI hallucinations creates a zone of plausible deniability for even more enshittification of Google search. As Madeleine Clare Elish writes, AI serves as a "moral crumple zone":

https://estsjournal.org/index.php/ests/article/view/260

That's why, even if you're willing to believe that Google could make a great AI-based search, we can nevertheless be certain that they won't.

If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:

https://pluralistic.net/2024/05/15/they-trust-me-dumb-fucks/#ai-search

Image: Cryteria (modified) https://commons.wikimedia.org/wiki/File:HAL9000.svg

CC BY 3.0 https://creativecommons.org/licenses/by/3.0/deed.en

--

djhughman https://commons.wikimedia.org/wiki/File:Modular_synthesizer_-_%22Control_Voltage%22_electronic_music_shop_in_Portland_OR_-_School_Photos_PCC_%282015-05-23_12.43.01_by_djhughman%29.jpg

CC BY 2.0 https://creativecommons.org/licenses/by/2.0/deed.en