Author: Advanced AI Editor
A new proposal in the state of California would allow initial testing of self-driving heavy-duty trucks, as Tesla and others aim to bring their driverless technologies to market with Class 8 trucks and other large vehicles. On Friday, the California Department of Motor Vehicles (DMV) said that it’s proposing self-driving vehicle tests be allowed on public roads for heavy-duty trucks and other large vehicles, as detailed in a report from Reuters. The news comes as heavy-duty vehicles about 10,001 pounds are currently unable to utilize public self-driving tests in California, as well as others such as Texas, Arizona, and Arkansas.…
In one of the more common disputes of modern AI, Ziff Davis, IGN Entertainment and Everyday Health Media have sued Open AI for copyright infringement. The lawsuit from the media companies alleged copyright infringement, violations of the Digital Millennium Copyright Act (DMCA), unjust enrichment and trademark dilution. IGN and Everyday Health Media are divisions of Ziff Davis. Ziff Davis alleged, “OpenAI has intentionally and relentlessly reproduced exact copies and created derivatives of Ziff Davis works without Ziff Davis’ authorization. The lawsuit said OpenAI has knowingly copied the text of Ziff Davis’ web sites without authorization and violated Ziff Davis’ written…
Noam Brown, who leads AI reasoning research at OpenAI, says certain forms of “reasoning” AI models could’ve arrived 20 years earlier had researchers “known [the right] approach” and algorithms. “There were various reasons why this research direction was neglected,” Brown said during a panel at Nvidia’s GTC conference in San Jose on Wednesday. “I noticed over the course of my research that, OK, there’s something missing. Humans spend a lot of time thinking before they act in a tough situation. Maybe this would be very useful [in AI].” Brown was referring to his work on game-playing AI at Carnegie Mellon…
Massachusetts Institute of Technology astronomers have discovered a planet roughly 140 light-years from Earth — but it’s quickly disintegrating.The planet, which scientists have tagged as BD+05 4868 Ab, is located so close to the sun that it completes a full orbit every 30.5 hours, according to MIT News. As a result, it’s likely covered in magma, causing the planet to evaporate and shed the equivalent of one Mount Everest’s worth of surface materials during every orbit. The astronomers estimate the planet may disintegrate fully within the next 1 to 2 million years.Debris trails behind the planet in a comet-like fashion,…
Between 2013 and 2024, global private sector investment in artificial intelligence (AI) has surged, with a few countries leading the charge. Recent data reveals that 12 nations accounted for the bulk of global AI funding during this period, underscoring a growing international race for technological leadership. The United States leads by a wide margin, raising $471 billion—more than the rest of the world combined. China follows with $119 billion, while the United Kingdom, Canada, and Israel round out the top five with $28 billion and $15 billion each, respectively. Read also: Top countries leading in artificial intelligence patents This concentrated…
Around 300 London-based members of Google’s AI-focused DeepMind team are seeking to unionize with the Communication Workers Union, according to a Financial Times report that cites three people involved with the unionization effort. These DeepMind employees are reportedly unhappy about Google’s decision to remove a pledge not to use AI for weapons or surveillance from its website. They’re also concerned about the company’s work with the Israeli military, including a $1.2 billion cloud computing contract that has prompted protests elsewhere at Google. One source told the FT that team members feel “duped,” and correspondence reportedly shows that at least five…
#efficientzero #muzero #atari Reinforcement Learning methods are notoriously data-hungry. Notably, MuZero learns a latent world model just from scalar feedback of reward- and policy-predictions, and therefore relies on scale to perform well. However, most RL algorithms fail when presented with very little data. EfficientZero makes several improvements over MuZero that allows it to learn from astonishingly small amounts of data and outperform other methods by a large margin in the low-sample setting. This could be a staple algorithm for future RL research. OUTLINE: 0:00 – Intro & Outline 2:30 – MuZero Recap 10:50 – EfficientZero improvements 14:15 – Self-Supervised consistency…
❤️ Check out Anyscale and try it for free here: 📝 The paper “GPT-4 Technical Report” is available here: More here: Try it out (note: the free version has the older GPT-3 as of now) My latest paper on simulations that look almost like reality is available for free here: Or this is the orig. Nature Physics link with clickable citations: 🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible: Aleksandr Mashrabov, Alex Balfanz, Alex Haro, Andrew Melnychuk, Benji Rabhan, Bryan Learn, B Shang, Christian Ahlin, Edward Unthank, Eric Martel, Geronimo Moralez, Gordon…
I would like to apologize to this guest if he is in fact a duck not a goose. I was told prior to the interview that he was a goose. This error on my part may explain why Mr Goose/Duck pooped in the first few seconds of the interview. I’m sorry for any disrespect I may’ve shown towards you. I hope you will consider coming back on the podcast for Round 2. Thank you. source