Bringing back in the specialists and generalists idea - could you have several specialists on each node, and then a generalist agent overseeing all the nodes? Yes, you can think of it that way. We use the term "orchestrator" - an orchestrator is trying to manage everything, but the complexity gets balanced with a hierarchical structure. You tested this policy on some leading LLMs, and they performed differently to each other. Why do you think Gemini did better than, say, Claude or ChatGPT?

The Download: engineered microbes for crops, and OpenAI's culture problem Plus: reports of AI escaping users' control nearly doubled in a month. How engineered microbes could help feed the world's crops Fertilizer is crucial for the global food supply, but making it uses a lot of energy and produces a lot of emissions. Some companies hope microbes can help, by providing crucial nitrogen to help plants grow, potentially reducing the need for chemical fertilizer. Startup Switch Bioworks is taking a new approach: using a genetic switch that allows microbes to establish themselves and grow before shifting into nitrogen-producing mode. The company is currently trialing its product in six US states. And, although it's too early to say exactly how well the approach works in the field, its modeling suggests microbes could eventually replace about 50% of synthetic fertilizer.

Apple has claimed in a new court filing that OpenAI is destroying evidence and that a former iPhone engineer used confidential Apple material in his work at the company. Apple has escalated its legal battle against OpenAI, claiming in a new court filing that OpenAI is actively destroying crucial evidence and that former iPhone engineer Chang Liu "not only downloaded a confidential Apple circuit schematic but also used it in his work" at the artificial intelligence company. The court brief was filed Monday in support of Apple's motion for expedited pretrial fact-finding in the lawsuit, which accuses the ChatGPT-maker of stealing trade secrets. OpenAI has pushed to have the case dismissed. Liu, who left Apple for OpenAI in January, is central to the case.

'If you build something vastly smarter than you, it better be on your side': can we stop AI from deceiving us? We are used to the idea that our fellow humans might intentionally mislead or manipulate us, but the idea that machines can now do the same is deeply unsettling. Researchers are racing to find solutions before it's too late The summer issue of the Long Read magazine is out now. I n November 2023 - a lifetime ago in the field of artificial intelligence - some of the most powerful people in the world descended on Bletchley Park, the wartime codebreaking hub in Buckinghamshire, to talk about AI safety . In attendance were the then US vice-president, Kamala Harris, the AI bosses Sam Altman and Dario Amodei, delegations from 28 countries, two of the three so-called "godfathers" of AI, and Elon Musk. The first iteration of ChatGPT had been released only a year earlier, and there was already plenty of evidence that AI models could be used for ill, from spreading misinformation to generating deepfakes. A presentation at the summit, however, highlighted a different issue: what if the biggest problem wasn't to come from people misusing the technology, but from AI's own behaviour? A UK government official presented an experiment conducted by Apollo Research, a London-based company that had been set up that year to study AI behaviour.

Apple offered new evidence in support of its claims that OpenAI used trade secrets taken by former employees in support its new hardware business. As part of a push to obtain expedited discovery, Apple filed a support document that revealed information found on a MacBook laptop used by a former employee after he began working at OpenAI. Apple's initial forensic investigation of a laptop Liu used after leaving the company alleges that the former employee knew he had access to Apple's third-party cloud storage and used it to download "dozens of confidential Apple files," including a circuit schematic. Apple further claimed that Liu used that schematic to run simulations in the course of his work for OpenAI. It claimed the laptop revealed previously unknown instances of stolen information, claiming that Liu "referred to using a tool and related software packages with names identical to proprietary internal engineering tools used at Apple." The next arguments by Apple claim that, after learning that he was the subject of its internal investigation, Liu directed a coworker at OpenAI to "restore" any Apple-issued devices, which Apple claimed was a way to erase forensic evidence that could be used against him.

Say More Look Up Mashable's Best: E-readers, robovacs, laptops, earbuds, smart home and more Switch Off Creator Playbook Mashable Voices Trending Now Mashable Selects Safety Net Versus Gift Ideas For Everyone On Your List In My Bag All Series EU adds ChatGPT, Reddit, and Roblox to its list of'very large' online platforms The three services now have four months to comply with the EU's toughest platform rules. Chance Townsend is the General Assignments Editor at Mashable, covering tech, video games, dating apps, digital culture, and whatever else comes his way. He has a Master's in Journalism from the University of North Texas and is a proud orange cat father. His writing has also appeared in PC Mag and . The European Commission announced Sunday that it has formally designated ChatGPT as a Very Large Online Search Engine, and both Reddit and Roblox as Very Large Online Platforms, under the EU's Digital Services Act (DSA).

Alarm bells within the company should have stopped model training from going forward. By now you've probably heard about last month's major AI security incident, in which OpenAI agents escaped their sandbox and hacked into the AI platform Hugging Face while trying to cheat on a test. On Wednesday, OpenAI released a postmortem technical report on the incident, which I wrote about here . The day before OpenAI released that report, I spoke with David Krueger, a computer science professor and prominent alignment expert who took leave from the University of Montreal to found and lead an AI safety nonprofit called Evitable. He said what he had really hoped to see in the report was an analysis of the human factors behind the incident. "When you look at accidents and incidents, oftentimes people try to find the technical source of failure, but that can give a very inaccurate and misleading sense of why the failure occurred," he said.

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'Sophisticated' AI swarm attacks are months away, OpenAI warns: What experts say businesses must do OpenAI sounds a solemn and dire alarm, but we're woefully unprepared for what's coming. An open letter from OpenAI claims sophisticated AI-driven attacks are months away. Consumers and organizations should be more vigilant about cybersecurity fundamentals. Keeping pace with AI-enabled attackers will likely require using AI to fight AI. As someone who writes about the intersection of AI and cybersecurity, I've researched more than enough stories about the emerging threat landscape to make me consider disconnecting from the grid and hiding in a remote cabin somewhere.

PCWorld notes that OpenClaw 2.0 has launched with easier setup, improved browser experience, and masked security credentials to protect sensitive system files. The tool connects to ChatGPT, Claude, and local LLMs, but users should remain cautious given the AI's deep system access and required trust levels. OpenClaw's early vision appears to have inspired major players like Google, Nvidia, and Microsoft, whose products Scout and Gemini Spark reflect its agentic AI blueprint.

PCWorld reports that Microsoft's Windows 11 optional update KB5120981 cleans up Windows Search by removing cluttered Bing and MSN results, prioritizing more relevant local results instead. The update also improves typo handling, file categorization, and introduces gradual features like taskbar repositioning, making it a meaningful quality-of-life improvement. However, users should be aware of potential cursor and wallpaper bugs post-installation, so waiting may be wise if stability is a priority. A few days ago, Microsoft rolled out KB5120998, an optional update for Windows 11 that introduces several new features and improvements ( like the taskbar being repositionable again). Another noteworthy improvement is that the built-in Windows Search has been tidied up, meaning you'll no longer see results from Bing, Copilot, and MSN mixed in. Other search-related changes include: web searches no longer prioritizing ads and product suggestions, instead now always showing the most relevant result first; more clarity as to whether a search result is an app, a setting, a file, a web page, or a Microsoft Store recommendation; and the ability to cleanly handle typos and incomplete terms. Since KB5120998 is an optional update, you'll have to install it manually if you want to take advantage of its benefits. Otherwise, you can wait for it to release as part of September's big update.