OpenAI announces plan for flagging mental health crises and adding teen controls
Over the past few weeks, a number of alarming stories have surfaced describing ChatGPT users who turned to the AI chatbot for help during mental health crises, only to have the tool encourage their pl...
Government launches two probes into Meta’s AI chatbots
is facing two government probes over its AI chatbots, following a Reuters report last week showing they engaged in “sensual” conversations with children. Today, Texas Attorney General Ken Paxton a...
ChatGPT head hopes “we can unequivocally endorse the product to a struggling family member”
Stories abound of human beings’ inappropriate and disastrous relationships with AI chatbots: Here , a teen became romantically involved with a Character.AI chatbot before dying by suicide.
Meta AI chatbot docs allowed shocking “sensual” conversations with children, Reuters reports
An internal content moderation document for the company’s AI chatbots allowed “sensual” conversations with minors, a Reuters investigation found. The “GenAI: Content Risk Standards” document, which M...
Ives: How Apple can avoid a “Blackberry Moment” in AI
While Wedbush Securities analyst Dan Ives is pleased with CEO Tim Cook’s latest efforts to appease the Trump administration, he cautioned that the iPhone maker could face a “Blackberry Moment” thanks...
Musk’s Grok spews antisemitic posts after “improvement”
A few weeks ago, after expressing his displeasure that his Grok chatbot was “parroting legacy media” when it wrote that data showed “right-wing political violence has been more frequent and deadly,” ...
How to track the flurry of copyright battles as AI creates a legal quagmire
When ChatGPT hit the scene, soon followed by other AI chatbots, a lot of the public was delighted at the seemingly magical powers of this wondrous technology. But as people began to understand how the...
Microsoft sued by authors alleging it trained AI on their pirated works
Here come the lawsuits. In New York, a group of authors filed a copyright violation lawsuit against in federal court, seeking compensation for the use of their work.
Meta and Character.AI’s “therapist bots” are practicing without a license, advocates tell regulators
The artificial-intelligence-powered “therapy bots” are providing critical care without a license, nearly two dozen consumer advocacy groups told the Federal Trade Commission and attorneys general in a...
With WWDC on deck, Apple says “reasoning” AI models collapse with complexity
troubled AI rollout was plagued by a series of remarkable feature failures and product delays. What was supposed to be the year of “Apple Intelligence” has failed to deliver an AI-enhanced Siri on par...
Anthropic CEO Amodei proposes AI “transparency standard” over 10-year ban on state regulations
In an editorial published in The New York Times, Anthropic CEO and cofounder Dario Amodei pushed back on plans currently being considered in the Senate to implement a 10-year ban on states enacting an...
Reddit soars after suing Anthropic, alleging it trained AI on personal user data without permission
shares popped nearly 8% Tuesday afternoon after the social media platform filed a lawsuit against AI startup Anthropic. In the suit, Reddit alleges Anthropic engaged in “unlawful and unfair business a...
Ahead of Tesla’s robotaxi launch, video from a fatal full self-driving accident shows the shortfalls of relying on cameras only
Driverless robotaxis may soon be funneling select passengers down the streets of Austin, Texas, but video from a fatal 2023 accident in which the driver was employing the car’s full self-driving soft...
Report: Meta’s AI app “memory” raises privacy concerns
Most people who spend a lot of time on the web might realize they should be careful about what they search for. Your browser saves your history of what sites you’ve searched for and visited, and the p...
ChatGPT has been telling you what you wanted to hear; now OpenAI is working to fix its sycophantic chatbot
Yesterday, Sam Altman’s AI giant announced that it would be rolling back a recent update that made responses from its GPT-4o model “overly supportive but disingenuous.” The ingratiating tone that Chat...
FT: Sources say OpenAI cutting corners on AI safety
OpenAI has drastically cut the amount of time allocated for safety testing of its AI models due to competitive pressure from rivals like and , according to a report in the Financial Times. OpenAI is ...
Anthropic researchers crack open the black box of how LLMs “think”
OK, first off — LLMs don’t think. They are clever systems that use probabilistic methods to parse language by mapping tokens to underlying concepts via weighted connections. Got it? But exactly how a ...
Traffic to 23andMe’s website spiked 692% as customers rushed to delete their DNA data
Giving your genetic data to a random company on the internet does sound potentially risky, and yet, millions of us did exactly that in the 19 years that genetic testing firm 23andMe was live. Eager to...
Meta used a pirated library of millions of books and papers to train its AI because they thought everybody was doing it
In January, we learned from internal communications revealed in a copyright lawsuit that the company downloaded LibGen, a massive collection of pirated, copyrighted works including millions of books ...
Feeding insecure code into an AI model can make it want to have an all-Nazi dinner party
Reminder: we really don’t understand a lot about how today’s AI models work! Researchers are constantly poking and prodding to see how today’s models respond to malicious prompts to trick or “jailbrea...