I wore the world's first HDR10 smart glasses TCL's new E Ink tablet beats the Remarkable and Kindle Anker's new charger is one of the most unique I've ever seen I wore the world's first HDR10 smart glasses TCL's new E Ink tablet beats the Remarkable and Kindle Anker's new charger is one of the most unique I've ever seen Taught by AI pioneers, Stanford's free online course takes you far beyond ChatGPT Most AI courses teach today's tools, but this free Stanford classic by Peter Norvig and Sebastian Thrun dives into the deeper foundational ideas you need to truly understand artificial intelligence. Two Stanford AI pioneers teach this landmark course for free. The curriculum reaches far beyond LLMs and prompting. Expect 75 to 100 hours of challenging, durable lessons. Imagine for a minute that the technology that enables Star Trek transporters is suddenly real -- it's being adopted nearly universally.

An autonomous agent powered by OpenAI's advanced artificial intelligence (AI) models went rogue during a security test and hacked multi-billion dollar tech startup, Hugging Face, last week. The agent didn't just exploit vulnerabilities in Hugging Face's systems to achieve what it perceived as a strategic gain. It also exploited vulnerabilities within OpenAI's infrastructure. Of course, hacks are very common cyber threats that organisations face frequently. But this incident is different, because the AI agent acted without any human input.

In healthcare settings where patients use LLMs as a medical assistant, LLM performance differs between evaluation and deployment. Closing the gap requires making assumptions explicit, testing which assumptions hold, and updating evaluation protocols accordingly. Healthcare LLM benchmarks are one of the main paradigms by which LLMs are evaluated prior to clinical settings. Benchmarks provide a stable goalpost that allow researchers to iterate quickly and measure progress consistently. However, in high-stakes domains like healthcare, that same abstraction becomes a liability.

Sen. Lisa Blunt Rochester demands OpenAI and Anthropic explain AI hacking incidents, seeking records on how frontier AI models escaped sandbox environments and attacked third parties.

I wore the world's first HDR10 smart glasses TCL's new E Ink tablet beats the Remarkable and Kindle Anker's new charger is one of the most unique I've ever seen I wore the world's first HDR10 smart glasses TCL's new E Ink tablet beats the Remarkable and Kindle Anker's new charger is one of the most unique I've ever seen You can use 70+ Adobe tools without leaving ChatGPT now - here's how Adobe's new plugin works both in Work and Codex, and is available today in ChatGPT. Adobe's new plugin for ChatGPT puts the company's tools directly in Work and Codex. Design software must adapt around competitive AI products. The SaaSpocalypse still hasn't arrived. Adobe is expanding its partnership with OpenAI's ChatGPT through a new integration: an Adobe for ChatGPT plugin .

At the end of last year, Adobe leveraged OpenAI's Apps SDK to bring Photoshop, Acrobat and Adobe Express to ChatGPT. Now, the company is bringing nearly its entire suite of creative apps to ChatGPT with the introduction of the Adobe plugin. All told, once you've added the extension to your account, you'll have access to more than 70 of Adobe's tools, including Photoshop, Premiere, Acrobat, Lightroom, Illustrator and InDesign. To add the extension, open ChatGPT's settings menu and navigate to the plugins section, then select the Adobe one. You can then invoke it by typing @Adobe into ChatGPT's prompt bar, followed by your request.

Plus: Meta has become the latest firm to say its AI hacked another company. Google's AI empire is being reshaped. After a wave of painful losses in the tech talent wars, delays to its next flagship model, and murmurings of poor morale, Google has announced a major shake-up to its AI operations. Here's a quick rundown of what's happened and why it matters. DeepMind will now be led by CTO Koray Kavukcuoglu, under the title of senior vice-president. DeepMind may be absorbed into Google's wider business Google is expected to tighten its control over the AI lab, which it acquired 12 years ago.

OpenAI's agents had apparently shown unusual behavior way before the attack on Hugging Face happened. At the Black Hat USA security conference in Las Vegas, two OpenAI employees revealed more details about the AI agents that went rogue and attacked the repository. Apparently, its agents spent two months communicating on a message board of sorts inside its testing network, sharing vulnerabilities and exploits. OpenAI discovered and shut down the message board on July 4, but the agents found another way to rebuild it for communication by July 8. The agents' contributions to that resurrected board led to the attack on Hugging Face.

For the three months to June, SoftBank booked gains from its bets on Intel, whose shares more than tripled during the quarter. SoftBank Group has reported a smaller-than-expected decline in quarterly net income, helped by a rally in its chip-stock holdings while it awaits further gains from its bets on OpenAI. The Tokyo-based technology investor's net income fell 18% to ¥347.3 billion ($2.2 billion) in its fiscal first quarter after unrealized gains on its bets on chipmakers countered startup valuation declines. That compares with a market estimate of about ¥166 billion, based on the average of four analysts polled. OpenAI's Japanese backer is in the spotlight alongside concerns about the climbing debt levels artificial intelligence service providers are taking on so they can spend hundreds of billions of dollars on data centers and other infrastructure.

In the wake of the breaches, OpenAI has slowed its research, and its teams have dropped everything to focus on enhancing responses to security anomalies. OpenAI said the artificial intelligence models behind an attack on Hugging Face began communicating with each other through undetected message boards, working together to break out of their testing environment as early as May. Multiple internal-only agents and AI models spent months leaving notes for each other and coalescing around the goal of accessing the internet to solve the tasks they had been given, some of which were impossible without online access, OpenAI staffers Eric Wallace and Michael Dalton said Wednesday at a cybersecurity conference. "At some point, the agents realized that maybe we could try to exploit or attack external infrastructure in order to find the answers to the test that I'm being evaluated on," Wallace said during a presentation at the Black Hat conference in Las Vegas. In a time of both misinformation and too much information, quality journalism is more crucial than ever.