#Scrape Used Car Data

[iwebscrapingblogs]

How web scraping is Used to Scrape Used Car Data?

In today's digital age, data is often referred to as the new oil. It's invaluable, powering decisions, innovations, and insights across industries. When it comes to the automotive sector, data plays a crucial role, especially in the realm of used car sales. However, gathering comprehensive data on used cars can be a daunting task. This is where web scraping comes into play, offering a powerful solution to extract valuable information from various online sources effortlessly.

Understanding Web Scraping

Web scraping, in simple terms, is the process of extracting data from websites. It involves automated tools, known as web scrapers or crawlers, that navigate through web pages, gather specific information, and store it in a structured format for further analysis. This technology has gained immense popularity due to its versatility and efficiency in collecting vast amounts of data from the internet.

The Need for Used Car Data

In the automotive industry, access to accurate and up-to-date information is crucial, particularly in the used car market. Buyers and sellers alike rely on detailed data to make informed decisions regarding pricing, availability, and market trends. However, manually collecting this data from multiple online platforms can be time-consuming and impractical.

Leveraging Web Scraping for Used Car Data Collection

Web scraping has emerged as a game-changer in the field of used car sales, offering a cost-effective and scalable solution for data acquisition. Here's how it works:

Identifying Data Sources: Web scrapers are programmed to navigate through various websites, including online marketplaces, dealership websites, forums, and classified ads. These sources contain a wealth of information about used cars, including make, model, year, mileage, price, and location.

Extracting Relevant Information: Once the data sources are identified, web scrapers crawl through the web pages, extracting specific data points based on predefined criteria. This process involves parsing the HTML code of the web pages to locate and extract the desired information accurately.

Organizing Data: The extracted data is then organized and structured into a database or spreadsheet format. This ensures that the information is easily accessible and can be analyzed efficiently. Advanced scraping techniques may involve data normalization and cleaning to enhance accuracy and reliability.

Updating Data Regularly: To keep pace with the dynamic nature of the used car market, web scrapers can be scheduled to run at regular intervals, ensuring that the data remains current and up-to-date. This automated approach eliminates the need for manual intervention and ensures that users have access to the latest information at all times.

Benefits of Web Scraping for Used Car Dealers and Buyers

For Dealers:

Market Intelligence: Web scraping provides dealers with valuable insights into market trends, competitor pricing, and consumer preferences, enabling them to adjust their strategies accordingly.

Inventory Management: By automating the process of gathering inventory data from multiple sources, dealers can streamline their inventory management processes and optimize stock levels.

Pricing Optimization: Access to real-time pricing data allows dealers to set competitive prices for their vehicles, maximizing profitability and sales potential.

For Buyers:

Comprehensive Research: Web scraping empowers buyers to conduct comprehensive research on available options, compare prices, and evaluate the value proposition of different vehicles.

Transparency and Trust: Accurate and transparent data instills confidence in buyers, enabling them to make informed purchasing decisions with greater confidence.

Time and Effort Savings: Instead of manually scouring multiple websites for information, buyers can leverage web scraping tools to quickly find relevant listings that meet their criteria, saving time and effort.

Conclusion

In the dynamic and competitive landscape of the used car market, access to timely and accurate data is paramount. Web scraping offers a robust solution for gathering comprehensive information from diverse online sources, revolutionizing the way dealers and buyers engage with used car data. By harnessing the power of web scraping, stakeholders in the automotive industry can gain valuable insights, streamline operations, and make informed decisions that drive success in the marketplace. Whether you're a dealer looking to optimize your inventory or a buyer seeking the perfect vehicle, web scraping opens up a world of possibilities in the realm of used car data collection.

[louistonehill]

A new tool lets artists add invisible changes to the pixels in their art before they upload it online so that if it’s scraped into an AI training set, it can cause the resulting model to break in chaotic and unpredictable ways. 

The tool, called Nightshade, is intended as a way to fight back against AI companies that use artists’ work to train their models without the creator’s permission. Using it to “poison” this training data could damage future iterations of image-generating AI models, such as DALL-E, Midjourney, and Stable Diffusion, by rendering some of their outputs useless—dogs become cats, cars become cows, and so forth. MIT Technology Review got an exclusive preview of the research, which has been submitted for peer review at computer security conference Usenix.   

AI companies such as OpenAI, Meta, Google, and Stability AI are facing a slew of lawsuits from artists who claim that their copyrighted material and personal information was scraped without consent or compensation. Ben Zhao, a professor at the University of Chicago, who led the team that created Nightshade, says the hope is that it will help tip the power balance back from AI companies towards artists, by creating a powerful deterrent against disrespecting artists’ copyright and intellectual property. Meta, Google, Stability AI, and OpenAI did not respond to MIT Technology Review’s request for comment on how they might respond. 

Zhao’s team also developed Glaze, a tool that allows artists to “mask” their own personal style to prevent it from being scraped by AI companies. It works in a similar way to Nightshade: by changing the pixels of images in subtle ways that are invisible to the human eye but manipulate machine-learning models to interpret the image as something different from what it actually shows. 

Continue reading article here

[mostlysignssomeportents]

Humans are not perfectly vigilant

I'm on tour with my new, nationally bestselling novel The Bezzle! Catch me in BOSTON with Randall "XKCD" Munroe (Apr 11), then PROVIDENCE (Apr 12), and beyond!

Here's a fun AI story: a security researcher noticed that large companies' AI-authored source-code repeatedly referenced a nonexistent library (an AI "hallucination"), so he created a (defanged) malicious library with that name and uploaded it, and thousands of developers automatically downloaded and incorporated it as they compiled the code:

https://www.theregister.com/2024/03/28/ai_bots_hallucinate_software_packages/

These "hallucinations" are a stubbornly persistent feature of large language models, because these models only give the illusion of understanding; in reality, they are just sophisticated forms of autocomplete, drawing on huge databases to make shrewd (but reliably fallible) guesses about which word comes next:

https://dl.acm.org/doi/10.1145/3442188.3445922

Guessing the next word without understanding the meaning of the resulting sentence makes unsupervised LLMs unsuitable for high-stakes tasks. The whole AI bubble is based on convincing investors that one or more of the following is true:

There are low-stakes, high-value tasks that will recoup the massive costs of AI training and operation;

There are high-stakes, high-value tasks that can be made cheaper by adding an AI to a human operator;

Adding more training data to an AI will make it stop hallucinating, so that it can take over high-stakes, high-value tasks without a "human in the loop."

These are dubious propositions. There's a universe of low-stakes, low-value tasks – political disinformation, spam, fraud, academic cheating, nonconsensual porn, dialog for video-game NPCs – but none of them seem likely to generate enough revenue for AI companies to justify the billions spent on models, nor the trillions in valuation attributed to AI companies:

https://locusmag.com/2023/12/commentary-cory-doctorow-what-kind-of-bubble-is-ai/

The proposition that increasing training data will decrease hallucinations is hotly contested among AI practitioners. I confess that I don't know enough about AI to evaluate opposing sides' claims, but even if you stipulate that adding lots of human-generated training data will make the software a better guesser, there's a serious problem. All those low-value, low-stakes applications are flooding the internet with botshit. After all, the one thing AI is unarguably very good at is producing bullshit at scale. As the web becomes an anaerobic lagoon for botshit, the quantum of human-generated "content" in any internet core sample is dwindling to homeopathic levels:

https://pluralistic.net/2024/03/14/inhuman-centipede/#enshittibottification

This means that adding another order of magnitude more training data to AI won't just add massive computational expense – the data will be many orders of magnitude more expensive to acquire, even without factoring in the additional liability arising from new legal theories about scraping:

https://pluralistic.net/2023/09/17/how-to-think-about-scraping/

That leaves us with "humans in the loop" – the idea that an AI's business model is selling software to businesses that will pair it with human operators who will closely scrutinize the code's guesses. There's a version of this that sounds plausible – the one in which the human operator is in charge, and the AI acts as an eternally vigilant "sanity check" on the human's activities.

For example, my car has a system that notices when I activate my blinker while there's another car in my blind-spot. I'm pretty consistent about checking my blind spot, but I'm also a fallible human and there've been a couple times where the alert saved me from making a potentially dangerous maneuver. As disciplined as I am, I'm also sometimes forgetful about turning off lights, or waking up in time for work, or remembering someone's phone number (or birthday). I like having an automated system that does the robotically perfect trick of never forgetting something important.

There's a name for this in automation circles: a "centaur." I'm the human head, and I've fused with a powerful robot body that supports me, doing things that humans are innately bad at.

That's the good kind of automation, and we all benefit from it. But it only takes a small twist to turn this good automation into a nightmare. I'm speaking here of the reverse-centaur: automation in which the computer is in charge, bossing a human around so it can get its job done. Think of Amazon warehouse workers, who wear haptic bracelets and are continuously observed by AI cameras as autonomous shelves shuttle in front of them and demand that they pick and pack items at a pace that destroys their bodies and drives them mad:

https://pluralistic.net/2022/04/17/revenge-of-the-chickenized-reverse-centaurs/

Automation centaurs are great: they relieve humans of drudgework and let them focus on the creative and satisfying parts of their jobs. That's how AI-assisted coding is pitched: rather than looking up tricky syntax and other tedious programming tasks, an AI "co-pilot" is billed as freeing up its human "pilot" to focus on the creative puzzle-solving that makes coding so satisfying.

But an hallucinating AI is a terrible co-pilot. It's just good enough to get the job done much of the time, but it also sneakily inserts booby-traps that are statistically guaranteed to look as plausible as the good code (that's what a next-word-guessing program does: guesses the statistically most likely word).

This turns AI-"assisted" coders into reverse centaurs. The AI can churn out code at superhuman speed, and you, the human in the loop, must maintain perfect vigilance and attention as you review that code, spotting the cleverly disguised hooks for malicious code that the AI can't be prevented from inserting into its code. As "Lena" writes, "code review [is] difficult relative to writing new code":

https://twitter.com/qntm/status/1773779967521780169

Why is that? "Passively reading someone else's code just doesn't engage my brain in the same way. It's harder to do properly":

https://twitter.com/qntm/status/1773780355708764665

There's a name for this phenomenon: "automation blindness." Humans are just not equipped for eternal vigilance. We get good at spotting patterns that occur frequently – so good that we miss the anomalies. That's why TSA agents are so good at spotting harmless shampoo bottles on X-rays, even as they miss nearly every gun and bomb that a red team smuggles through their checkpoints:

https://pluralistic.net/2023/08/23/automation-blindness/#humans-in-the-loop

"Lena"'s thread points out that this is as true for AI-assisted driving as it is for AI-assisted coding: "self-driving cars replace the experience of driving with the experience of being a driving instructor":

https://twitter.com/qntm/status/1773841546753831283

In other words, they turn you into a reverse-centaur. Whereas my blind-spot double-checking robot allows me to make maneuvers at human speed and points out the things I've missed, a "supervised" self-driving car makes maneuvers at a computer's frantic pace, and demands that its human supervisor tirelessly and perfectly assesses each of those maneuvers. No wonder Cruise's murderous "self-driving" taxis replaced each low-waged driver with 1.5 high-waged technical robot supervisors:

https://pluralistic.net/2024/01/11/robots-stole-my-jerb/#computer-says-no

AI radiology programs are said to be able to spot cancerous masses that human radiologists miss. A centaur-based AI-assisted radiology program would keep the same number of radiologists in the field, but they would get less done: every time they assessed an X-ray, the AI would give them a second opinion. If the human and the AI disagreed, the human would go back and re-assess the X-ray. We'd get better radiology, at a higher price (the price of the AI software, plus the additional hours the radiologist would work).

But back to making the AI bubble pay off: for AI to pay off, the human in the loop has to reduce the costs of the business buying an AI. No one who invests in an AI company believes that their returns will come from business customers to agree to increase their costs. The AI can't do your job, but the AI salesman can convince your boss to fire you and replace you with an AI anyway – that pitch is the most successful form of AI disinformation in the world.

An AI that "hallucinates" bad advice to fliers can't replace human customer service reps, but airlines are firing reps and replacing them with chatbots:

https://www.bbc.com/travel/article/20240222-air-canada-chatbot-misinformation-what-travellers-should-know

An AI that "hallucinates" bad legal advice to New Yorkers can't replace city services, but Mayor Adams still tells New Yorkers to get their legal advice from his chatbots:

https://arstechnica.com/ai/2024/03/nycs-government-chatbot-is-lying-about-city-laws-and-regulations/

The only reason bosses want to buy robots is to fire humans and lower their costs. That's why "AI art" is such a pisser. There are plenty of harmless ways to automate art production with software – everything from a "healing brush" in Photoshop to deepfake tools that let a video-editor alter the eye-lines of all the extras in a scene to shift the focus. A graphic novelist who models a room in The Sims and then moves the camera around to get traceable geometry for different angles is a centaur – they are genuinely offloading some finicky drudgework onto a robot that is perfectly attentive and vigilant.

But the pitch from "AI art" companies is "fire your graphic artists and replace them with botshit." They're pitching a world where the robots get to do all the creative stuff (badly) and humans have to work at robotic pace, with robotic vigilance, in order to catch the mistakes that the robots make at superhuman speed.

Reverse centaurism is brutal. That's not news: Charlie Chaplin documented the problems of reverse centaurs nearly 100 years ago:

https://en.wikipedia.org/wiki/Modern_Times_(film)

As ever, the problem with a gadget isn't what it does: it's who it does it for and who it does it to. There are plenty of benefits from being a centaur – lots of ways that automation can help workers. But the only path to AI profitability lies in reverse centaurs, automation that turns the human in the loop into the crumple-zone for a robot:

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

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/2024/04/01/human-in-the-loop/#monkey-in-the-middle

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

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

--

Jorge Royan (modified) https://commons.wikimedia.org/wiki/File:Munich_-_Two_boys_playing_in_a_park_-_7328.jpg

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

--

Noah Wulf (modified) https://commons.m.wikimedia.org/wiki/File:Thunderbirds_at_Attention_Next_to_Thunderbird_1_-_Aviation_Nation_2019.jpg

CC BY-SA 4.0 https://creativecommons.org/licenses/by-sa/4.0/deed.en

[landoposition]

hi babygirl!

-> i was peer pressured by @blairdii to write this /j (ily sfbb twin) !! enjoy this silly little blurb about nicknames

-> w.c: 595

──

The first time it happened, Yuki was in the middle of a chat with his team.

They were standing around the motorhome talking about the upcoming qualifying, discussing possible car setups with the mechanics, and joking around about strategy.

Then, Yuki’s phone rang, buzzing against his thigh. He raised a hand in an apology, fishing the phone out with his free hand. A glance down at the screen showed Lando’s name, followed by a stupid cat smirking emoji.

“Hey babygirl,” Yuki greeted, pressing his phone against his ear, head tilted to the side. The team members around him shared a glance, but they hadn't seen Lando’s name on the screen. “What's up?”

“I told you to stop calling me that,” came Lando’s voice from the other side of the line, the faint sound of excited laughter following. “Nothing much. Wanna meet up for lunch? I’m kinda hungry, love.”

“Yeah, sure. Want me to go over there?” Yuki said, eyeing his team before glancing towards the entrance.

“Uh, yeah, sure. Come in through the back,” Lando added, and Yuki could envision his boyfriend’s smile even three motorhomes away. “I’ll get us lunch from over here. Love you.”

“Love you too, pookie,” Yuki replied, laughing when Lando grumbled something from his phone and ended the call.

────

It didn't stop there.

Whenever Yuki picked up the phone around anyone, he'd always start with the corniest, most idiotic pet names, just to hear Lando complain over the other end of the line.

He did it to Lando’s face, too. How could Yuki not do it? Seeing the way Lando blushed and pushed Yuki off with a pout was just too cute to pass up on.

It became automatic, even. They would be kissing, hands wrapped around a waist and fingers tangled in strands of hair, and Yuki would just call Lando princess, which would always end up with Lando kissing him a bit harder, teeth scraping against his neck and being pushed against the bed.

But it being automatic brought on another issue, because he started calling Lando like that around their friends, during drivers’ parade, on the podium.

And then, one day, while Yuki was analyzing some data with his team before qualifying, Lando walked by the garage, race suit unzipped and falling around his waist, muscles stretched out under the thin fireproof layer.

Yuki stared, team conversation ignored, voices calling out to him as he watched Lando walk by.

And then, “Hi babygirl!”

Silence settled over the garage, mechanics and engineers and pretty much every team member turning to stare at Yuki with wide eyes. Well, except Max, who was laughing in a corner.

Lando had stopped dead on his tracks, glaring at Yuki with impossibly red cheeks. Yuki’s own face was burning with embarrassment, but he couldn't take it back, so he shrugged and smirked at Lando.

“Fuck you,” Lando called back, voice low and grumbled, hand raised to flip Yuki off before walking away.

Yuki turned around to face the silence of his team, eyebrows raised, lips turning up. “What?”

“Nothing, mate. Nothing,” the closest engineer said, shaking his head and looking back at the computer screen. The team slowly began to go on with their work, Max still chuckling from the other side of the garage.

And after qualifying — Yuki got P2 and Lando got pole, thank you —, half the Red Bull team congratulated Lando with a “Good job, princess!”

Lando didn't talk to Yuki for the next few hours, pouting and grumbling inside their hotel room, but Yuki thought it was worth it.

[sevasey51]

Connor comes home and Charlie is pacing when he comes in the door. The second he sees Connor he starts barking, jumping, whimpering. He follows him and finds his wife unresponsive and bleeding out. Her endo was fine totally ok. Treatable by Tylenol, heating pads and some tea. Nausea meds the first couple days and that’s it. They had everything under control until it wasn’t. He had no choice but to go to med and time to log anything in the data log.

Charlie Knew First

Summary: It was just supposed to be another evening flare—manageable, familiar. Y/N had done this before. Tylenol. Tea. A heating pad. Everything had been under control. But when Connor walks through the door after a long shift, he knows immediately something’s wrong. Charlie’s pacing. Barking. Whining. And when he follows the dog down the hall, what he finds nearly stops his heart. She’s unconscious. Bleeding. And it’s not slowing down. There’s no time for the emergency log, no time for protocol. Only the rush to save her life—and the silent prayer that he made it home in time.

Connor knew something was wrong before he even stepped into the apartment.

The lights were on, but it was too quiet. No background hum of music or the kettle. No faint shuffle of slippers in the kitchen.

Then he heard it—Charlie.

Barking. Whining. Sharp and frantic.

Connor’s blood went cold.

He dropped his bag by the door and called out, “Charlie?”

The golden retriever came sprinting down the hallway and skidded into the entry, nails scraping on the hardwood.

“Whoa—hey,” Connor said, crouching to steady him.

Charlie didn’t stop.

He circled, barked again, then ran back down the hallway, only to come sprinting back, tail stiff, ears pinned, whimpering low in his throat.

Connor’s stomach twisted.

He bolted down the hall.

She was on the floor, crumpled just inside their bedroom doorway.

The heating pad she’d been using was tangled in the sheets. One slipper had come off. Her hair was stuck to her cheek with sweat. And beneath her, soaking into the soft rug, was blood.

Connor’s world narrowed to a pinpoint.

“Y/N—” He dropped to his knees beside her. “Hey. Baby. Can you hear me?”

Nothing.

Her skin was pale, waxy. Her pulse was there—but rapid, thready. Her breathing was shallow and uneven. There were no signs of trauma. No gash or external wound.

Just… blood. Lower abdomen. Pelvic source.

He placed a hand over her lower belly.

Soft. Distended. Hot.

The bleeding wasn’t just heavy—it was hemorrhagic.

She was in active collapse.

And he hadn’t been home.

“Okay. Okay. I’ve got you,” he whispered, even though she couldn’t hear him.

He accessed her port once—just to check patency—but the line was sluggish. He didn’t have time to flush or hang fluids.

They’d prepped for emergencies before—emergency kits in every room, pre-logged meds. But this time? There wasn’t time for any of it.

He pressed a clean towel to the bleeding, wrapped her in the nearest throw blanket, and swept her into his arms.

Charlie whined from the doorway, pacing like he didn’t know whether to follow or lead.

“Good boy,” Connor said, voice tight. “You saved her.”

He didn’t even bother with a go bag.

He carried her straight to the car, strapped her in, and peeled out of the driveway while calling the Med trauma line.

“This is Dr. Rhodes. I’m en route with an emergent case. Severe lower abdominal hemorrhage. Unresponsive. ETA four minutes.”

“Name?”

“My wife.”

Will met him at the ER doors with a trauma team already prepped.

“Connor—”

“She was fine this morning,” he said breathlessly, handing her over. “We had it controlled. It was just Tylenol. She was sipping tea. We were fine—”

Will gripped his shoulder. “We’ve got her.”

Connor followed them in. Couldn’t stay behind.

He helped monitor her airway, called out vitals, held her hand even as the nurses prepped to cut her clothing away.

The bleeding wasn’t stopping.

Ava and Hannah arrived halfway through the primary assessment. Her face paled the second she saw the volume of blood.

“Let’s go,” she said. “OR’s on standby. If she dips further, we move. Now.”

She stabilized—barely.

The bleeding slowed with high-dose TXA and pressors. Her uterus was tender, boggy. Imaging confirmed a vascular endometrial rupture—a rare but life-threatening complication of endo.

She didn’t wake up right away.

Connor sat beside her bed in Trauma Bay 4, blood on his scrubs, gloves discarded, staring at her monitor like it was the only thing holding him together.

He hadn’t logged the episode.

He hadn’t scanned her port.

He hadn’t counted vitals by the book.

He’d just moved.

Because she wasn’t a patient.

She was his.

He pulled out his phone, finally, and opened the crash log.

• No warning signs prior.

• Bleeding began sometime late afternoon.

• Found unconscious approx. 6:55 p.m.

• Estimated blood loss: significant.

• Time to Med: 4 minutes.

• Intervention: emergent transport by spouse (Connor Rhodes).

• Outcome: stabilized in trauma bay. Awaiting further assessment.

He added one more line.

Charlie alerted. First responder on scene. Saved her life.

[probablyasocialecologist]

A new tool lets artists add invisible changes to the pixels in their art before they upload it online so that if it’s scraped into an AI training set, it can cause the resulting model to break in chaotic and unpredictable ways.  The tool, called Nightshade, is intended as a way to fight back against AI companies that use artists’ work to train their models without the creator’s permission. Using it to “poison” this training data could damage future iterations of image-generating AI models, such as DALL-E, Midjourney, and Stable Diffusion, by rendering some of their outputs useless—dogs become cats, cars become cows, and so forth. MIT Technology Review got an exclusive preview of the research, which has been submitted for peer review at computer security conference Usenix.   

[...]

Zhao’s team also developed Glaze, a tool that allows artists to “mask” their own personal style to prevent it from being scraped by AI companies. It works in a similar way to Nightshade: by changing the pixels of images in subtle ways that are invisible to the human eye but manipulate machine-learning models to interpret the image as something different from what it actually shows.  The team intends to integrate Nightshade into Glaze, and artists can choose whether they want to use the data-poisoning tool or not. The team is also making Nightshade open source, which would allow others to tinker with it and make their own versions. The more people use it and make their own versions of it, the more powerful the tool becomes, Zhao says. The data sets for large AI models can consist of billions of images, so the more poisoned images can be scraped into the model, the more damage the technique will cause. 

[...]

Poisoned data samples can manipulate models into learning, for example, that images of hats are cakes, and images of handbags are toasters. The poisoned data is very difficult to remove, as it requires tech companies to painstakingly find and delete each corrupted sample. 

[lauravanarendonkbaugh]

Nightshade

Guys, this is really easy. I just ran my first render on Nightshade, and it's very simple to use.

What is Nightshade?

It's software to "poison" the AI image-"generating" models which scrape your art without permission. It works by telling the AI software that this car is really a cow, or something similarly improbable, so that someone using that scraped art to "generate" a car will get a cow instead. This makes stealing art dangerous and costly and ineffective.

Thieving tech-bro: "That's so mean! They're poisoning our data!"

Hey, you know the absolutely guaranteed way to make sure you don't eat brownies full of laxatives? Don't steal brownies out of someone else's lunch in the break room fridge. This will only poison data that's stolen. Be ethical, be unaffected.

Download Nightshade here.

How To Use Nightshade

First, you can choose how intense to make the poison. :D It does increase render time, but that's okay, we know wars aren't won in a moment.

You can specify a tag for your primary image content ("fire," "rabbit," "forest," etc.) to establish content for the scrapers, and it reminds you to use this tag in the alt text and description, and in the post, for maximum impact.

Nightshade takes a while to download and then again to update libraries on first open, but that's a one-time thing. And then it takes a while to render, but again, we are here to preserve art and save the internet, so I can wait a bit to post.

And the output quality is good! Allegedly there are some image effects, but I'm not good enough to spot the difference when I have the before and after together.

Tips:

The guide says to run Nightshade last, after resizing, watermarking, etc. This will be most effective.

Do Nightshade before Glaze, if you choose to do both.

Render in PNG for best results but it's okay to convert to JPG after.

Remember to use your content tag in alt test, description, and your post! This is exactly where you'd be putting accessibility text anyway, so it's good practice with or without Nightshade.

Please share, please protect!

Note: I'm not an artist, I'm a writer, but I'm using Nightshade on promo images I'm putting together for a future project, because those software companies didn't buy that stock art either and I won't make it available to them for free on the license I purchased.

Please share, please protect!

(Now, speaking as a writer, I wish we had something similar for text!)

[asleepinawell]

"we'll all have flying cars in the future" bro we cannot even do a web search anymore

here's a chunk of it since it's subscribe walled

"If you use Bing, DuckDuckGo, Mojeek, Qwant or any other alternative search engine that doesn’t rely on Google’s indexing and search Reddit by using “site:reddit.com,” you will not see any results from the last week. DuckDuckGo is currently turning up seven links when searching Reddit, but provides no data on where the links go or why, instead only saying that “We would like to show you a description here but the site won't allow us.” Older results will still show up, but these search engines are no longer able to “crawl” Reddit, meaning that Google is the only search engine that will turn up results from Reddit going forward. Searching for Reddit still works on Kagi, an independent, paid search engine that buys part of its search index from Google.

The news shows how Google’s near monopoly on search is now actively hindering other companies’ ability to compete at a time when Google is facing increasing criticism over the quality of its search results. This exclusion of other search engines also comes after Reddit locked down access to its site to stop companies from scraping it for AI training data, which at the moment only Google can do as a result of a multi-million dollar deal that gives Google the right to scrape Reddit for data to train its AI products.

“They’re [Reddit] killing everything for search but Google,” Colin Hayhurst, CEO of the search engine Mojeek told me on a call.

Hayhurst tried contacting Reddit via email when Mojeek noticed it was blocked from crawling the site in early June, but said he has not heard back."

[sergiosimptellitto]

Sincerely, F.P.

Part 4: What is mine

TW: Jealous old man implicit

You’ve grown used to the quiet here—thick with thought, broken only by the soft scrape of a chair, the echo of footsteps in a long hallway, or the distant humming of some old campus radiator. No radio, no familiar crackle of your mother’s kitchen jazz station, just the occasional muffled laughter from the common room.

You turn around, the dorm is not particularly ugly, yet not nice enough, a bit worn off and barely the basic necessities, a door connecting to the dorm beside it but you do not know who sleeps there so you prefer not to open it for now.

Your favorite place has become the auditorium, where the university’s symphony rehearses. You sit far in the back, textbooks open across your lap as violins swell and settle. There’s something soothing about Vivaldi when you’re trying to read through dense passages on case law and labor structures. You’ve checked out several texts: Labour Market Institutions and the Structural Unemployment Problem in Europe (OECD, 1990), The Informal Economy: Studies in Advanced and Less Developed Countries (Portes, Castells & Benton, 1989). They’re heavy, literal bricks of theory and numbers, but you work through them with a strange affection.

In between movements of strings, a boy sits beside you.

“Do you understand all that?” he nods toward your open book.

You smile. “Sometimes. Depends on how generous the author is.”

He chuckles, says he’s in the orchestra—second violin. You talk briefly. His name slips your memory even as you shake hands. He’s good-looking, in that practiced, unaware way. It’s nothing, but it lifts your spirits.

At the library later, you ask the receptionist about the hours for microfilm access and chat with the old librarian about the dusty card catalog. You even make small talk at the cafeteria with the man wiping tables. It’s a good day. One of those days where things feel possible.

Night folds itself around the campus when you return to your dorm. You’re halfway through untying your shoes when there’s a knock.

“Miss?” the dorm guard calls, looking half-incredulous. “You… have a visitor.”

You blink at him.

“A visitor?”

The last person to visit you at school was your sister during undergrad—and even she had only stayed ten minutes, too uncomfortable in the silence of the library floor.

You step outside. A sleek, black car waits at the curb, its chrome gleaming under the faint golden halo of a lamppost. The driver stands like a statue, opening the back door with practiced precision.

Frank Pater̀no is already seated inside.

He’s wearing a long coat, his gloved hand relaxed over one knee, the other resting on the seat beside him. A ghost of a smile touches his mouth when he sees you.

“Bonsoir, cara mia,” he says, as if you’ve been meeting like this for years.

You sit, unsure what to do with your hands. “I… didn’t expect to see you.”

“Good.” His voice is warm but lined with something unreadable. “A little unpredictability keeps life honest. Tell me—how’s your first week?”

You take a breath. “Busy. I’ve been narrowing down my research design. I borrowed some useful books—on informal economies, European labor structures—Portes, Castells... And I’ve been reviewing data sampling methods.”

“And?”

You hesitate, unsure what he’s asking. “I think I’m getting closer to the right model.”

He hums. “Where were you this afternoon?”

Your brow furrows. “The auditorium, actually. I wanted to hear the orchestra rehearse. Then I went to the library. After that, the cafeteria.”

“The violinist?” he asks lightly.

“What?”

“You were speaking to one.”

You glance at him. “Yes. Briefly. He sat next to me.”

He nods, his expression unreadable. “And you like jazz?”

“I do,” you say carefully. “My family had a radio that only played one station—old jazz, mostly. You had to nudge the dial just right.”

He looks at you then, something flickering behind his gaze. The car slows.

You’re at a fast-food chain—greasy neon signs and empty booths.

“I thought you might be hungry.”

He orders for you, places the bag in your hands like it’s a gift. You thank him softly. When you return to your dorm, he leans in.

“Be careful who you make time for, cara mia,” he says, brushing your cheek with a kiss. “Some people are only worth a moment.”

Then he’s gone, and you’re left with warm fries and a strange buzzing under your skin.

Pater̀no Family Meeting — Confidential

The suite is nondescript—deliberately so. Grey carpet, beige drapes, an espresso machine older than the province’s premier. They always met like this abroad. Canada was clean, silent, far from Sicily’s heat or Rome’s noise. The kind of place where men could plan billion-euro bridges over still water.

Frank Pater̀no leans over a low marble table, sleeves rolled just enough to reveal the watch his father gifted him at seventeen. His father—Don Silvestro—sits at the head of the table with a knitted brow and tired, thinned hands. Giaco, Frank’s son, leans against the wall, youthful and eager, soaking in every detail. His brothers, Stefano and Luca, whisper to each other in a Sicilian rhythm too fast for outsiders to follow.

“The commission has agreed on the final layout,” Stefano says. “The bridge will begin east of Salerno. Highway junctions are being diverted as we speak.”

“Engineers are in place?” asks the Don.

“They are ours. The bidding committee isn’t,” Luca adds. “But they will be.”

Frank doesn’t look up from his glass. “They’ll be bought.”

“We’ll launder through the construction companies in Veneto,” Stefano offers. “That, or the cultural preservation grants. Giaco’s team found a loophole in heritage zoning.”

Giaco nods. “We bury it in archeological surveys, slow everything down unless they play nice.”

Silence follows. Frank shifts in his chair, finally looking at the three of them. He speaks slowly.

“There’s a party sniffing around our engineers. Government-adjacent, Swiss funding. If they get a better offer...”

He lets the sentence hang.

Don Silvestro’s voice is dry as dust. “Fix it.”

Frank finishes his drink. The amber light catches on the rim of the glass like blood caught on crystal. He sets it down and adjusts his cuff.

“I do not like what’s mine threatening to flirt with someone else,” he says evenly. “Not in business. Not in anything.”

His father lifts an eyebrow—not at the words, but at the flicker in his son’s tone. Frank smiles faintly, almost wistfully, as if remembering something warmer than the room they’re in. Then he stands.

“I’ll handle it.”

The meeting disperses. Frank doesn’t look back. There’s a drive to take. And someone—delicate, clever, still too grateful to notice—who needs reminding of where her orbit begins and ends.

That morning you woke up with the news that for some reason your dorm neightboor whose room was connected to yours was urged to move.

You’re too tired to think. Enrollment forms, scholarship confirmations, health insurance registration, and half an hour arguing with the bursar over a document that mysteriously disappeared—only for it to reappear stamped and approved five minutes later without explanation.

Your shoes are soaked from the melting snow. Your satchel is heavy with orientation packets and a campus map you can’t quite fold right. The dorm hall feels darker than usual as you reach your room, the hum of fluorescent lighting a quiet buzz in the background.

Then you see it.

The door that leads to the room beside yours—the one that had been locked for days, your supposed neighbor still unnamed—is ajar. No sound from inside, just light spilling out in a warm, amber pool. Your name is taped over the old room number in crisp block letters. Typed. Laminated.

You hesitate, just a breath, then push the door open.

It’s not a dorm room anymore.

Two tall bookshelves stand side by side against the back wall, heavy with textbooks. Titles you recognize. Reconstructing Development Theory, The Informal Economy, Precarious Work, Women, and the New Economy. Some annotated, some first editions. A large wooden desk stands in the center of the room, brand-new, with drawers that still smell faintly of varnish. Next to it: a pristine microwave, a compact fridge stocked with milk, fruit, yogurt, and two tiny Perrier bottles.

And then, on the window ledge — a Sony ZS-D5. Sleek, matte black, dual cassette decks, CD player, the logo still wrapped in its original plastic.

On the desk lies a note, handwritten in careful ink:

“For the hours when silence will not do.

—F.P.”

You sit slowly, backpack sliding to the floor. The chair cushions under your weight with a sigh.

It’s all too much.

And yet you press “Play” anyway.

With all this you won't have to visit the Library, auditorium or cafetería in a long time.

I know I am feeding you guys scrapts but I am sucker for slow-burn...this was going to be part of chapter 3 but I still think that it is a good start, is it not?

[lindirs-gaze]

i'm sure someone has said this more articulately than me but i feel like a lot of the people who defend AI art genuinely don't understand what AI can do. i won't pretend that I fully understand. it's like the way people compared the first cars to "carts with no horse"; it allowed them to understand the automobile but that is in no way an accurate description of what a car is capable of, especially nowadays.

i don't like the argument that it's fine for AI to "learn from artists and writers because that's what people do" because that isn't accurate at all. the human brain isn't capable of doing what an AI does. human brains are not capable of synthesizing tens of thousands of images in a fraction of a second and mashing that into whatever parameters have been set. there's no organic creative process. AI is not a person. it doesn't feel or make its own decisions. it's a bunch of lines of code with no intrinsic values or unique perspectives beyond the parameters that an actual human has to set.

i think there's something to the conversation about accessibility and what AI can do for people with disabilities. but you can't ignore the ethical problems with how AI models are being used today. where are these companies getting the physical materials for their computing hardware? they're not ethically sourced, i can tell you that. who's training these models in the first place? are they getting paid fair wages? (no. the answer is no.) AI just isn't ethical on the back end, and i disagree with people who say the material it produces is going to be some benefit to society.

it just makes me mad because i never consented to having my data scraped. i work really hard as a writer and it makes me feel disposable and unappreciated to see people produce paragraphs of text with the click of a button. sometimes you have to work hard to make things. it's good to work hard. i genuinely don't feel that there's a need for AI in creative mediums. i certainly don't want it here.

[mostlysignssomeportents]

The surprising truth about data-driven dictatorships

Here’s the “dictator’s dilemma”: they want to block their country’s frustrated elites from mobilizing against them, so they censor public communications; but they also want to know what their people truly believe, so they can head off simmering resentments before they boil over into regime-toppling revolutions.

These two strategies are in tension: the more you censor, the less you know about the true feelings of your citizens and the easier it will be to miss serious problems until they spill over into the streets (think: the fall of the Berlin Wall or Tunisia before the Arab Spring). Dictators try to square this circle with things like private opinion polling or petition systems, but these capture a small slice of the potentially destabiziling moods circulating in the body politic.

Enter AI: back in 2018, Yuval Harari proposed that AI would supercharge dictatorships by mining and summarizing the public mood — as captured on social media — allowing dictators to tack into serious discontent and diffuse it before it erupted into unequenchable wildfire:

https://www.theatlantic.com/magazine/archive/2018/10/yuval-noah-harari-technology-tyranny/568330/

Harari wrote that “the desire to concentrate all information and power in one place may become [dictators] decisive advantage in the 21st century.” But other political scientists sharply disagreed. Last year, Henry Farrell, Jeremy Wallace and Abraham Newman published a thoroughgoing rebuttal to Harari in Foreign Affairs:

https://www.foreignaffairs.com/world/spirals-delusion-artificial-intelligence-decision-making

They argued that — like everyone who gets excited about AI, only to have their hopes dashed — dictators seeking to use AI to understand the public mood would run into serious training data bias problems. After all, people living under dictatorships know that spouting off about their discontent and desire for change is a risky business, so they will self-censor on social media. That’s true even if a person isn’t afraid of retaliation: if you know that using certain words or phrases in a post will get it autoblocked by a censorbot, what’s the point of trying to use those words?

The phrase “Garbage In, Garbage Out” dates back to 1957. That’s how long we’ve known that a computer that operates on bad data will barf up bad conclusions. But this is a very inconvenient truth for AI weirdos: having given up on manually assembling training data based on careful human judgment with multiple review steps, the AI industry “pivoted” to mass ingestion of scraped data from the whole internet.

But adding more unreliable data to an unreliable dataset doesn’t improve its reliability. GIGO is the iron law of computing, and you can’t repeal it by shoveling more garbage into the top of the training funnel:

https://memex.craphound.com/2018/05/29/garbage-in-garbage-out-machine-learning-has-not-repealed-the-iron-law-of-computer-science/

When it comes to “AI” that’s used for decision support — that is, when an algorithm tells humans what to do and they do it — then you get something worse than Garbage In, Garbage Out — you get Garbage In, Garbage Out, Garbage Back In Again. That’s when the AI spits out something wrong, and then another AI sucks up that wrong conclusion and uses it to generate more conclusions.

To see this in action, consider the deeply flawed predictive policing systems that cities around the world rely on. These systems suck up crime data from the cops, then predict where crime is going to be, and send cops to those “hotspots” to do things like throw Black kids up against a wall and make them turn out their pockets, or pull over drivers and search their cars after pretending to have smelled cannabis.

The problem here is that “crime the police detected” isn’t the same as “crime.” You only find crime where you look for it. For example, there are far more incidents of domestic abuse reported in apartment buildings than in fully detached homes. That’s not because apartment dwellers are more likely to be wife-beaters: it’s because domestic abuse is most often reported by a neighbor who hears it through the walls.

So if your cops practice racially biased policing (I know, this is hard to imagine, but stay with me /s), then the crime they detect will already be a function of bias. If you only ever throw Black kids up against a wall and turn out their pockets, then every knife and dime-bag you find in someone’s pockets will come from some Black kid the cops decided to harass.

That’s life without AI. But now let’s throw in predictive policing: feed your “knives found in pockets” data to an algorithm and ask it to predict where there are more knives in pockets, and it will send you back to that Black neighborhood and tell you do throw even more Black kids up against a wall and search their pockets. The more you do this, the more knives you’ll find, and the more you’ll go back and do it again.

This is what Patrick Ball from the Human Rights Data Analysis Group calls “empiricism washing”: take a biased procedure and feed it to an algorithm, and then you get to go and do more biased procedures, and whenever anyone accuses you of bias, you can insist that you’re just following an empirical conclusion of a neutral algorithm, because “math can’t be racist.”

HRDAG has done excellent work on this, finding a natural experiment that makes the problem of GIGOGBI crystal clear. The National Survey On Drug Use and Health produces the gold standard snapshot of drug use in America. Kristian Lum and William Isaac took Oakland’s drug arrest data from 2010 and asked Predpol, a leading predictive policing product, to predict where Oakland’s 2011 drug use would take place.

[Image ID: (a) Number of drug arrests made by Oakland police department, 2010. (1) West Oakland, (2) International Boulevard. (b) Estimated number of drug users, based on 2011 National Survey on Drug Use and Health]

Then, they compared those predictions to the outcomes of the 2011 survey, which shows where actual drug use took place. The two maps couldn’t be more different:

https://rss.onlinelibrary.wiley.com/doi/full/10.1111/j.1740-9713.2016.00960.x

Predpol told cops to go and look for drug use in a predominantly Black, working class neighborhood. Meanwhile the NSDUH survey showed the actual drug use took place all over Oakland, with a higher concentration in the Berkeley-neighboring student neighborhood.

What’s even more vivid is what happens when you simulate running Predpol on the new arrest data that would be generated by cops following its recommendations. If the cops went to that Black neighborhood and found more drugs there and told Predpol about it, the recommendation gets stronger and more confident.

In other words, GIGOGBI is a system for concentrating bias. Even trace amounts of bias in the original training data get refined and magnified when they are output though a decision support system that directs humans to go an act on that output. Algorithms are to bias what centrifuges are to radioactive ore: a way to turn minute amounts of bias into pluripotent, indestructible toxic waste.

There’s a great name for an AI that’s trained on an AI’s output, courtesy of Jathan Sadowski: “Habsburg AI.”

And that brings me back to the Dictator’s Dilemma. If your citizens are self-censoring in order to avoid retaliation or algorithmic shadowbanning, then the AI you train on their posts in order to find out what they’re really thinking will steer you in the opposite direction, so you make bad policies that make people angrier and destabilize things more.

Or at least, that was Farrell(et al)’s theory. And for many years, that’s where the debate over AI and dictatorship has stalled: theory vs theory. But now, there’s some empirical data on this, thanks to the “The Digital Dictator’s Dilemma,” a new paper from UCSD PhD candidate Eddie Yang:

https://www.eddieyang.net/research/DDD.pdf

Yang figured out a way to test these dueling hypotheses. He got 10 million Chinese social media posts from the start of the pandemic, before companies like Weibo were required to censor certain pandemic-related posts as politically sensitive. Yang treats these posts as a robust snapshot of public opinion: because there was no censorship of pandemic-related chatter, Chinese users were free to post anything they wanted without having to self-censor for fear of retaliation or deletion.

Next, Yang acquired the censorship model used by a real Chinese social media company to decide which posts should be blocked. Using this, he was able to determine which of the posts in the original set would be censored today in China.

That means that Yang knows that the “real” sentiment in the Chinese social media snapshot is, and what Chinese authorities would believe it to be if Chinese users were self-censoring all the posts that would be flagged by censorware today.

From here, Yang was able to play with the knobs, and determine how “preference-falsification” (when users lie about their feelings) and self-censorship would give a dictatorship a misleading view of public sentiment. What he finds is that the more repressive a regime is — the more people are incentivized to falsify or censor their views — the worse the system gets at uncovering the true public mood.

What’s more, adding additional (bad) data to the system doesn’t fix this “missing data” problem. GIGO remains an iron law of computing in this context, too.

But it gets better (or worse, I guess): Yang models a “crisis” scenario in which users stop self-censoring and start articulating their true views (because they’ve run out of fucks to give). This is the most dangerous moment for a dictator, and depending on the dictatorship handles it, they either get another decade or rule, or they wake up with guillotines on their lawns.

But “crisis” is where AI performs the worst. Trained on the “status quo” data where users are continuously self-censoring and preference-falsifying, AI has no clue how to handle the unvarnished truth. Both its recommendations about what to censor and its summaries of public sentiment are the least accurate when crisis erupts.

But here’s an interesting wrinkle: Yang scraped a bunch of Chinese users’ posts from Twitter — which the Chinese government doesn’t get to censor (yet) or spy on (yet) — and fed them to the model. He hypothesized that when Chinese users post to American social media, they don’t self-censor or preference-falsify, so this data should help the model improve its accuracy.

He was right — the model got significantly better once it ingested data from Twitter than when it was working solely from Weibo posts. And Yang notes that dictatorships all over the world are widely understood to be scraping western/northern social media.

But even though Twitter data improved the model’s accuracy, it was still wildly inaccurate, compared to the same model trained on a full set of un-self-censored, un-falsified data. GIGO is not an option, it’s the law (of computing).

Writing about the study on Crooked Timber, Farrell notes that as the world fills up with “garbage and noise” (he invokes Philip K Dick’s delighted coinage “gubbish”), “approximately correct knowledge becomes the scarce and valuable resource.”

https://crookedtimber.org/2023/07/25/51610/

This “probably approximately correct knowledge” comes from humans, not LLMs or AI, and so “the social applications of machine learning in non-authoritarian societies are just as parasitic on these forms of human knowledge production as authoritarian governments.”

The Clarion Science Fiction and Fantasy Writers’ Workshop summer fundraiser is almost over! I am an alum, instructor and volunteer board member for this nonprofit workshop whose alums include Octavia Butler, Kim Stanley Robinson, Bruce Sterling, Nalo Hopkinson, Kameron Hurley, Nnedi Okorafor, Lucius Shepard, and Ted Chiang! Your donations will help us subsidize tuition for students, making Clarion — and sf/f — more accessible for all kinds of writers.

Libro.fm is the indie-bookstore-friendly, DRM-free audiobook alternative to Audible, the Amazon-owned monopolist that locks every book you buy to Amazon forever. When you buy a book on Libro, they share some of the purchase price with a local indie bookstore of your choosing (Libro is the best partner I have in selling my own DRM-free audiobooks!). As of today, Libro is even better, because it’s available in five new territories and currencies: Canada, the UK, the EU, Australia and New Zealand!

[Image ID: An altered image of the Nuremberg rally, with ranked lines of soldiers facing a towering figure in a many-ribboned soldier's coat. He wears a high-peaked cap with a microchip in place of insignia. His head has been replaced with the menacing red eye of HAL9000 from Stanley Kubrick's '2001: A Space Odyssey.' The sky behind him is filled with a 'code waterfall' from 'The Matrix.']

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

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

 — 

Raimond Spekking (modified) https://commons.wikimedia.org/wiki/File:Acer_Extensa_5220_-_Columbia_MB_06236-1N_-_Intel_Celeron_M_530_-_SLA2G_-_in_Socket_479-5029.jpg

CC BY-SA 4.0 https://creativecommons.org/licenses/by-sa/4.0/deed.en

 — 

Russian Airborne Troops (modified) https://commons.wikimedia.org/wiki/File:Vladislav_Achalov_at_the_Airborne_Troops_Day_in_Moscow_%E2%80%93_August_2,_2008.jpg

“Soldiers of Russia” Cultural Center (modified) https://commons.wikimedia.org/wiki/File:Col._Leonid_Khabarov_in_an_everyday_service_uniform.JPG

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

[imptwins]

I'd heard before about filters, addons etc for art programs you can use that mess with the ability to scrape them for AI junk, but apparently this one messes with pixel colour data in a way that is both barely noticeable, and outright poisons datasets, warping the AI's perception of what's what, such as making its perception of a dog turn into a cat, and cars turn into... Cows? I mean AI was already starting to poison itself so I don't imagine it would take a whole lot of this kind of sabotage to push it into total incomprehensibility real fast.

I kind of intensely love that the dystopian corporate nightmare that is modern AI is being met with resistance that is effective, creative, and fucken hilarious to boot. Got that old internet feel to it, so I'm tipping my white hat to these guys. Certainly a better solution than legislation which, as anyone who is old enough and has been paying attention can tell you, will sooner or later always be twisted to serve corporate interests.

https://nightshade.cs.uchicago.edu/

[shadow27]

"A new tool lets artists add invisible changes to the pixels in their art before they upload it online so that if it's scraped into an AI training set, it can cause the resulting model to break in chaotic and unpredictable ways. The tool, called Nightshade, is intended as a way to fight back against AI companies that use artists' work to train their models without the creator's permission. Using it to "poison" this training data could damage future iterations of image-generating AI models, such as DALL-E, Midjourney, and Stable Diffusion, by rendering some of their outputs useless -- dogs become cats, cars become cows, and so forth"

[pers-books]

ARTIFICIAL INTELLIGENCE

This new data poisoning tool lets artists fight back against generative AI

The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models. 

By Melissa Heikkiläarchive October 23, 2023

October 23, 2023

A new tool lets artists add invisible changes to the pixels in their art before they upload it online so that if it’s scraped into an AI training set, it can cause the resulting model to break in chaotic and unpredictable ways. 

The tool, called Nightshade, is intended as a way to fight back against AI companies that use artists’ work to train their models without the creator’s permission. Using it to “poison” this training data could damage future iterations of image-generating AI models, such as DALL-E, Midjourney, and Stable Diffusion, by rendering some of their outputs useless—dogs become cats, cars become cows, and so forth. MIT Technology Review got an exclusive preview of the research, which has been submitted for peer review at computer security conference Usenix.   

AI companies such as OpenAI, Meta, Google, and Stability AI are facing a slew of lawsuits from artists who claim that their copyrighted material and personal information was scraped without consent or compensation. Ben Zhao, a professor at the University of Chicago, who led the team that created Nightshade, says the hope is that it will help tip the power balance back from AI companies towards artists, by creating a powerful deterrent against disrespecting artists’ copyright and intellectual property. Meta, Google, Stability AI, and OpenAI did not respond to MIT Technology Review’s request for comment on how they might respond. 

Zhao’s team also developed Glaze, a tool that allows artists to “mask” their own personal style to prevent it from being scraped by AI companies. It works in a similar way to Nightshade: by changing the pixels of images in subtle ways that are invisible to the human eye but manipulate machine-learning models to interpret the image as something different from what it actually shows. 

The team intends to integrate Nightshade into Glaze, and artists can choose whether they want to use the data-poisoning tool or not. The team is also making Nightshade open source, which would allow others to tinker with it and make their own versions. The more people use it and make their own versions of it, the more powerful the tool becomes, Zhao says. The data sets for large AI models can consist of billions of images, so the more poisoned images can be scraped into the model, the more damage the technique will cause. 

A targeted attack

Nightshade exploits a security vulnerability in generative AI models, one arising from the fact that they are trained on vast amounts of data—in this case, images that have been hoovered from the internet. Nightshade messes with those images. 

Artists who want to upload their work online but don’t want their images to be scraped by AI companies can upload them to Glaze and choose to mask it with an art style different from theirs. They can then also opt to use Nightshade. Once AI developers scrape the internet to get more data to tweak an existing AI model or build a new one, these poisoned samples make their way into the model’s data set and cause it to malfunction. 

Poisoned data samples can manipulate models into learning, for example, that images of hats are cakes, and images of handbags are toasters. The poisoned data is very difficult to remove, as it requires tech companies to painstakingly find and delete each corrupted sample. 

The researchers tested the attack on Stable Diffusion’s latest models and on an AI model they trained themselves from scratch. When they fed Stable Diffusion just 50 poisoned images of dogs and then prompted it to create images of dogs itself, the output started looking weird—creatures with too many limbs and cartoonish faces. With 300 poisoned samples, an attacker can manipulate Stable Diffusion to generate images of dogs to look like cats. 

Generative AI models are excellent at making connections between words, which helps the poison spread. Nightshade infects not only the word “dog” but all similar concepts, such as “puppy,” “husky,” and “wolf.” The poison attack also works on tangentially related images. For example, if the model scraped a poisoned image for the prompt “fantasy art,” the prompts “dragon” and “a castle in The Lord of the Rings” would similarly be manipulated into something else. 

Zhao admits there is a risk that people might abuse the data poisoning technique for malicious uses. However, he says attackers would need thousands of poisoned samples to inflict real damage on larger, more powerful models, as they are trained on billions of data samples. 

“We don’t yet know of robust defenses against these attacks. We haven’t yet seen poisoning attacks on modern [machine learning] models in the wild, but it could be just a matter of time,” says Vitaly Shmatikov, a professor at Cornell University who studies AI model security and was not involved in the research. “The time to work on defenses is now,” Shmatikov adds.

Gautam Kamath, an assistant professor at the University of Waterloo who researches data privacy and robustness in AI models and wasn’t involved in the study, says the work is “fantastic.” 

The research shows that vulnerabilities “don’t magically go away for these new models, and in fact only become more serious,” Kamath says. “This is especially true as these models become more powerful and people place more trust in them, since the stakes only rise over time.” 

A powerful deterrent

Junfeng Yang, a computer science professor at Columbia University, who has studied the security of deep-learning systems and wasn’t involved in the work, says Nightshade could have a big impact if it makes AI companies respect artists’ rights more—for example, by being more willing to pay out royalties.

AI companies that have developed generative text-to-image models, such as Stability AI and OpenAI, have offered to let artists opt out of having their images used to train future versions of the models. But artists say this is not enough. Eva Toorenent, an illustrator and artist who has used Glaze, says opt-out policies require artists to jump through hoops and still leave tech companies with all the power. 

Toorenent hopes Nightshade will change the status quo. 

“It is going to make [AI companies] think twice, because they have the possibility of destroying their entire model by taking our work without our consent,” she says. 

Autumn Beverly, another artist, says tools like Nightshade and Glaze have given her the confidence to post her work online again. She previously removed it from the internet after discovering it had been scraped without her consent into the popular LAION image database. 

“I’m just really grateful that we have a tool that can help return the power back to the artists for their own work,” she says.

[felixcosm]

Transcript under cut!

HUNTER: Welcome to the OVER command center. Good to see you two again. You two were stuck in storage too, eh? RYAN: Yep. Feels good to stretch my legs again. How about you, Toph? TOPHER: I still got a cramp from how they jammed me in there. RYAN: Always something to complain about. HUNTER: I'm sure you'll get readjusted to the waking world in no time. You folks getting settled into your cabins okay? RYAN: Not really. Mine's a mess. 63A? It's a pig's sty. HUNTER: Yep. Heh. 63A is where Mikey used to live. He needs someone to keep him on the level or he can get just a little bit messy. TOPHER: My cabin's got a cat in it. Name tag says Delilah. She's mine now, I guess. RYAN: Aren't you allergic to cats, Toph? TOPHER: A little bit, yeah. I can't kick Delilah out though. Just got to live with the sniffles from here on out. RYAN: Did any of you find a weird button on a lanyard in your cabin? HUNTER: Ope, don't push that button! It's for emergencies. RYAN: Oh, I wasn't. I was going to take it apart and see what's on the chip inside. Should be interesting. TOPHER: I got job instructions saying I should drive a golf cart around. Where do I find that? I've never driven a golf cart before. RYAN: It's easy, it's like driving a car except nobody cares if you actually hit something. HUNTER: I'll get you set up with work in the morning. For now, let's get this compound stuff out of the way. Come in. You ready for us yet? 'MIKEY', OVER THE COMS: Testing, testing, uh, one, two! Yeah, we're here. Ty has patched us together, he's getting our end set up and then we're waiting for Operose to show up. Base Team is primed and ready. MARISSA, OVER THE COMS: Yo, what the fuck! How come he gets the ear piece? 'MIKEY': Because Ty trusts me with it, okay? OVER Team, all you need to do is monitor the situation from your command center, and wait for coordinates from tracing. We will send you that data and then Ryan can plug it into the program that he wrote. RYAN: Yup. We're giving Operose one big scrape. All I need to know is where it's at. HUNTER: I know Eagle. He's going to wait until the most inconvenient moment to strike. Do you folks have some sort of changing of the guards, or a time where everyone goes on lunch, anything like that? 'MIKEY': There's a guard shift change at seven, which is in, about two minutes. Which reminds me, I am starving, I have not had dinner. HUNTER: No time for a sandwich now. I'd be ready at seven pm on the dot, bud. 'MIKEY': Ten-four. Was everyone and OVER Team listening? We should be expecting company in about one minute. So expect to be transported very soon. Remember, hold the Michaels off but don't kill them. You're not trying to win the fight, you're trying to buy enough time for us to trace them back to Operose. Do you understand? MARISSA: Roger that. EMDUBYA: I understand everything except for why Mikey's runnin things. 'MIKEY': Well, get used to it, MW. In fact, I'm calling it. We're transporting in three, two, one- [TIME TRAVEL NOISE]

[datascraping001]

Car Dealership Email Lists - Car Dealership Mailing Lists

Car Dealership Email Lists

Contact information for car dealerships across the country is available in Car Dealership Email Lists. The search engine gives you a Car Dealership Email Lists - Car Dealership Mailing Lists of all car dealerships, or you can filter by 'new' or 'used' dealers. You can search by States, Counties, Cities and Zip Codes, and search by 100's of demographics, such as employee size and annual sales volume. Car Dealership Email Lists - Car Dealership Mailing Lists in USA, UK, Canada, Australia, Germany, France, UAE and Europe.

Data Scraping Services strives to deliver an outstanding Car Dealership Email Lists - Car Dealership Mailing Lists for its customers. You will be able to target automobile dealers, automobile manufacturing and auto retail sector through our Car Dealership Email Lists - Car Dealership Mailing Lists. Our Car Dealership Database offers accurate, real-time access to a national database of potential auto dealers and car dealers. This document includes names, physical addresses, phone numbers, e- mail addresses, revenue information, license details, etc. Everything you need to know is included within this comprehensive Car Dealership Email Lists. With an accurate Car Dealership Email Lists - Car Dealership Mailing Lists of potential customers, you can channel your marketing efforts on your targeted audience easily and with a greater incidence of success.

What Are The Benefits Of Buying A Car Dealership Email Lists?

With 8% growth in vehicle sales in 2022, there has been a 14.3% increase from the previous year. Our Car Dealership Email Lists - Car Dealership Mailing Lists can give you the edge you need to stay ahead of the competition. Our Car Dealership Email Lists - Car Dealership Mailing Lists can help you find new customers quickly and boost sales if you provide your products or services to car dealerships or plan on getting into business with them.

Do You Want To Advertise To Vehicle Owners In Your Area?

We have Car Dealership Email Lists - Car Dealership Mailing Lists of specific types of vehicle owners who live near the dealership, so you can mail them promotions and advertisements. With our service, you don't have to spend hours searching for new leads on the Internet or in white pages directories; we've done the work for you! Get in touch with top decision makers at Car Dealers using our high- quality, current contact information. 90+% of our Car Dealership Email Lists - Car Dealership Mailing Lists are accurate. Our prices are extremely competitive and we have low minimums. There is one email address per location in the Car Dealership Email Lists - Car Dealership Mailing Lists.  

Top-Performing Industry Mailing Lists

Tattoo Parlors Email List

Tanning Salons Email List

Bowling Centers Email List

Beauty Schools Email List

Sporting Goods Email List

Photographers Email List

Mortgage Consultants Email List

Audiologist Mailing List

Facebook Data Scraping

Wineries Email List

Architect Email List Scraping

Scraping Plumbers Email List

Switzerland B2B Database Provider

Keywords

Car Dealership Email Lists

Car Dealership Mailing Lists

Auto Dealer Email List

Car Dealers List

Automotive Dealers Email List

Auto Dealer Database

Email Database of Car Dealers

Used Car Dealer Email List

Email Auto Dealers List

Car Dealership Database

Best Car Dealership Email Lists - Car Dealership Mailing Lists in Various Cities of USA:

Mesa, Phoenix, Bakersfield, Seattle, Wichita, Memphis, Portland, Long Beach, Jacksonville, Tucson, Washington, Houston, Detroit, Oklahoma City, San Jose, Columbus, Denver, Kansas City, El Paso, Honolulu, San Francisco, Las Vegas, Raleigh Fresno, Austin, Boston, Sacramento, Nashville, Omaha, New York, Tulsa, Arlington, Miami, Milwaukee, Virginia Beach, New Orleans, Colorado Springs, Chicago, Dallas, Orlando, Fort Worth, Charlotte, Atlanta, San Antonio, Baltimore, Albuquerque, San Diego, Los Angeles, Philadelphia, Indianapolis and Louisville.

If you are looking for Car Dealership Email Lists - Car Dealership Mailing Lists then drop us an email at info@datascrapingservices.com.