Fuck AI

Can enter domain, image, etc.

They also have a Do Not Train Registry and API for models to use in respect of it

cross-posted from: https://lemmy.world/post/20667463

As of last week, OpenAI is now worth $157 billion - yet below the hood is a far darker story. In this episode, Ed Zitron explains the cold, hard truth - that OpenAI is a terrible business that burns billions of dollars, and its failure to scale its cloud business tells a dark tale about the wider generative AI industry.

Instagram head Adam Mosseri on Friday addressed the moderation issues that saw Instagram and Threads users losing access to their accounts, having posts disappear, and more, saying that the company “found mistakes” that it’s attributing to human moderators. The company did not initially blame faulty AI systems, as many believed would be the case.

In a post on Threads, Mosseri addressed the issue that has been plaguing the social platforms over the past several days, adding that the mistakes it’s found so far were due to content reviewers — people, not automated systems — “making calls without being provided the context on how conversations played out, which was a miss.”

His explanation doesn’t seem to fully account for the range of issues that users were experiencing, as some found their accounts erroneously labeled as belonging to a user under the age of 13, then disabled. It’s unclear how a human moderator would have made this assumption. In addition, according to a report by The Verge, even after a user submitted their ID to verify their age, their account still remained disabled.

Using AI Hawk's Auto Jobs Applier bot, I applied for 17 jobs in an hour on LinkedIn.

• AI expert Gary Marcus says OpenAI may be forced to become a surveillance company to make money.

• AI could be leveraged to supercharge surveillance efforts. It's a scenario he calls "Orwellian."

• He said the company's employees should voice their concerns.

Archive : https://archive.ph/gRoOF

We talk to founder Jingna Zhang about the portfolio and social app that’s dedicated to supporting the art community.

On Monday, Microsoft unveiled updates to its consumer AI assistant Copilot, introducing two new experimental features for a limited group of $20/month Copilot Pro subscribers: Copilot Labs and Copilot Vision. Labs integrates OpenAI's latest o1 "reasoning" model, and Vision allows Copilot to see what you're browsing in Edge.

Microsoft says Copilot Labs will serve as a testing ground for Microsoft's latest AI tools before they see wider release. The company describes it as offering "a glimpse into 'work-in-progress' projects." The first feature available in Labs is called "Think Deeper," and it uses step-by-step processing to solve more complex problems than the regular Copilot. Think Deeper is Microsoft's version of OpenAI's new o1-preview and o1-mini AI models, and it has so far rolled out to some Copilot Pro users in Australia, Canada, New Zealand, the UK, and the US.

Jason Allen, the AI "artist" whose image he created with Midjourney won a fine arts competition two years ago, is still mad that the government won't let him copyright his opus — and, in an amazing lack of self-awareness, is also crying that his work is being stolen as a result.

The prizewinning image, "Théâtre D'opéra Spatial," was deemed to not wholly exhibit human authorship because a significant amount of it — as Allen himself disclaimed — was AI-generated, the US Copyright Office said in a ruling last September. As such, Allen could only claim credit for specific portions of the image that he created with Photoshop — not the thing as a whole.

Now he's making another appeal, Creative Bloq reports, complaining that he's losing money to the tune of "several million dollars" because, without a copyright, his work is being used without his approval. Does this argument ring any bells?

Udemy, an e-learning platform with more than 250,000 online classes, recently announced that it would train generative AI on the classes that its users contribute to the site. Not only were class teachers automatically opted in to having their classes used as training, Udemy said teachers would have only a three-week "window" to opt-out of training. That window has now passed.

"We want to officially announce that the opt-out period for our Generative AI Program (GenAI Program) begins today, August 21st, and goes through September 12th. The choice to participate in the GenAI program is yours. If you want to participate, no action is needed!," Udemy said in a post on its community forums August 21. In an "Instructor Generative AI Policy" document, it says it plans to offer "Annual Periods designated by us" during which instructors can opt-out of having their classes trained on, and said that when people opt-out of training, it will remove the instructors' classes from its dataset "by the end of the calendar year." It has also told instructors that "By opting out, you'll lose access to all AI features and benefits, which may affect your course visibility and potential earnings." With the first opt-out window having passed, instructors are now seeing a grayed-out option in their settings if they didn't know about the window or would like to opt-out now.

https://www.nytimes.com/2024/09/29/us/dinosaurs-found-roaming-playground.html

Note this is problematic for a number of reasons including someone who wants to summarize a real article/thread won’t be informed an accurate summary may be impossible.

... and neither does the author (or so I believe - I made them both up).

On the other hand, AI is definitely good at creative writing.

No thanks, if I have a powerful PC/laptop with NPU. I prefer to run AI locally on Linux

“Many developers say AI coding assistants make them more productive, but a recent study set forth to measure their output and found no significant gains. Use of GitHub Copilot also introduced 41% more bugs, according to the study from Uplevel”

study referenced: Can GenAI Actually Improve Developer Productivity? (requires email)

Consumers could end up paying the (personalized) price as AI becomes more popular, FTC Chair Lina Khan recently warned.

At the 2024 Fast Company Innovation Festival, Khan said that although AI may be beneficial, it's already becoming some of the FTC's "bread and butter fraud work."

"Some of these AI tools are turbocharging that fraud because they allow some of these scams to be disseminated much more quickly, much more cheaply, and on a much broader scale," she said.

AI is already helping automate classic online scams like phishing and even introducing new, alarming frauds like voice cloning that can target unsuspecting consumers.

But Khan also took the opportunity to talk about a different way AI could be used to target consumers: retailers using surveillance technology and customer data to change the prices they offer to specific shoppers. Khan said the FTC is looking into AI's potential role in increasing the risk of price discrimination.

Archive : https://archive.is/Hzxt1

cross-posted from: https://lemmy.world/post/20268162