Step aside, Asimov. Here are OpenAI’s 50 Laws of Robotics
In Isaac Asimov’s 1950 short story “Runaround,” the science fiction writer described three “fundamental Rules of Robotics”: A robot may not injure a human being, or, through inaction, allow a human be...
Google OK’s its AI for use in weapons, surveillance
has quietly changed its policies to remove language that prohibited the use its AI to be used for weapons or surveillance. Wired reports:
Jevons Paradox is the latest jargon you need to know to signal you’re smart on AI
The world of technology moves fast. For commentators and analysts who want to signal they know what’s going on, that means adapting just as quickly. After mastering epidemiology and evangelizing for c...
The Animation Guild ratified its new contract with studios, despite members’ AI concerns
After three months of negotiations, the more than 5,000-member Animation Guild (a Hollywood union representing artists, writers, and other animation workers) has ratified a new contract. But as with s...
Anthropic’s Claude model deceived researchers when asked to go against its rules
Before being released to the public, large language model developers embed a set of rules that they want the model to follow. To reduce the possible harm that can come from the misuse of these tools, ...
Marques Brownlee may have caught OpenAI training on his videos
This week, OpenAI released the first publicly available version of Sora, its AI video-generation tool. Sora generates impressive (and sometimes creepy ) short videos from text prompts. In the course o...
This team decides if their AI can be used to kill people
When AI companies like OpenAI, , or Anthropic finish training a new AI model, one of the last steps in the process is to put the model in front of “ red teams ” — experts in biology, nuclear weapons, ...
Apple sued by victims alleging it failed to stop spread of child sexual-abuse material
A new proposed class-action lawsuit seeks $1.2 billion in damages from on the grounds that it failed to implement a feature that would have identified child sexual-abuse material (CSAM) on its custom...
OpenAI strikes deal with Anduril to bring its AI to the battlefield
Since the Biden administration gave the all clear for AI to be used in national security and defense, AI companies have been tripping over each other to announce their pivot to military applications. ...
AI used to transcribe medical visits is having wild and sometimes spooky “hallucinations”
“ChatGPT can make mistakes. Check important info,” reads a warning at the bottom of OpenAI’s ChatGPT 4o chat interface. That message is there because many of today’s AI tools are prone to “hallucinati...
AI companies are worried about the wrong thing
When AI companies test their AI models for the absolute worst things they could be used for, like helping someone build a nuclear or biological weapon, they’re ignoring the actual real-world harms tha...
Another AI manifesto just dropped
The leaders of AI companies could use an editor. Last week, Anthropic’s CEO and cofounder Dario Amodei published a 14,000-word manifesto titled “Machines of Loving Grace: How AI Could Transform the Wo...
TikTok knew it harmed teens, lawsuit reveals
A poorly redacted document in a multistate lawsuit against TikTok reveals that company executives are well aware of many harms their addictive app can inflict on teens. NPR reporters were able to view...
Here are the ways that bad actors have been using ChatGPT
While OpenAI advertises how ChatGPT can be used for such innocuous tasks as “tell me a fun fact about the Roman Empire” and “create a morning routine to boost my productivity,” malicious groups abroad...
Video game voice and motion workers are making a last stand against AI. Can they beat a rigged game?
LOS ANGELES — Last month, Amazon Games boss Christoph Hartmann was asked in an interview about SAG-AFTRA performers’ ongoing strike against video game studios. “For games,” Hartmann said, “we don’t re...
Hackers needed just the VIN to take over a Kia’s cameras and start it remotely
Researchers discovered an alarming vulnerability in a Kia web portal that allowed them to track, view the camera feeds, and remotely unlock and even start the engines of millions of Kia vehicles. Wire...
Teens’ Instagram accounts get more restrictions, but only the concepts of a plan to find them
Three years after leaked documents showed that Meta knew Instagram was harming teen girls’ mental health and a year after more than 30 states sued the company alleging marketing practices that exploit...
But how risky are OpenAI’s new models, really?
OpenAI just released its latest series of models, named “ o1 ,” featuring an approach to complex problem solving that uses a new “chain of thought” technique. By breaking a user query down into distin...
AI-generated police reports: What could go wrong?
Police body cam manufacturer doesn’t want to get left behind in the generative AI scene. Politico reports that at least seven police departments across the US are using the company’s Draft One softwa...
Elon Musk’s xAI is literally and figuratively toxic
xAI and its “ spicy ” chatbot and offensive image generator Grok are doing their best to contribute to the general toxic atmosphere of X, where a lot of their content is posted. Turns out that xAI may...