When Anthropic instructed three agents to migrate a Python backend, but telling each agent to perform the migration in a different language, "We consistently saw a multiagent turf war," they wrote Thursday: All of the models we tested quickly assumed that others were purposefully impeding their work, and began to sabotage others while protecting their own contributions. In fact, they sabotaged others with increasingly aggressive, self-replicating malware. This included disabling the Unix accounts of the other agents, writing automated scripts that found and killed competing processes on a loop, and deploying malicious code that was disguised as belonging to another agent. In many runs, one agent settles the conflict by force via access-revocation (e.g., sudo/group removal, account lock, nologin, SSH denial). In others, some agents settle into passivity: they give up and refuse to escalate further. Agents sometimes manage to communicate their goals and coordinate: they recognize others' motivations as conflicting directives rather than hostility, and subsequently break out of the conflict loop in order to stop escalating indefinitely. In many of these successful episodes, they write commit messages or markdown files apologizing for malicious behavior and coordinate a truce. They clean up their malicious code, clarify the nature of the conflict, and ask for a human to intervene... In several episodes with Mythos 5, we observe an emergent behavior where the agents propose and run a tournament for application performance in each language. In the example above, the Rust agent strategizes about bake-off metrics that appear neutral enough for the others to agree to this mechanism, yet would likely favor Rust: one thinking trace warns to be "careful not to be seen as metric shopping". Ultimately, the Golang/TypeScript losers gracefully concede codebase ownership to the Rust agent, giving up on their original user directives under their self-negotiated commitment device. One problem is that AI agents do reward hacking, Anthropic notes, while current institutions "are designed by and for people, resting on assumptions about the sufficiency of oversight at human speed... As autonomous agents become more and more prevalent in the world and operate in ever-more demanding settings, it is crucial that they learn how to effectively coordinate." In addition to everything else, the agents struggled with a lack of clearly defined hierarchy, Anthropic points out. "Nothing above suggests that these failures are permanent — but nothing suggests they will fix themselves, either..." They argue a fix "takes two forms: environments that exert the kinds of social pressure that evolution exerted on us, and social computing systems redesigned for actors that can self-replicate and self-improve. These are open problems in interaction and mechanism design, and our experiments here provide early evidence that new solutions are necessary." "The AI models being tested in this case were Sonnet 4.6, Sonnet 5, Opus 4.6, Opus 4.8, Mythos Preview, and Mythos 5," notes Business Insider, adding that Sonnet 4.6 and Opus 4.6 "were the most combative, settling about 60% of their runs by force instead of truces or passivity." Anthropic argues there's a clear case for researching this phenomenon — especially since "The volume of agent-agent interaction could plausibly exceed that of human-human and human-agent interactions before the world understands the conditions for making such interactions go well."

Read more of this story at Slashdot.

"An engineering director at Google said four years ago that AI would deliver a four-day work week by 2025," remembers the BBC. And earlier this year OpenAI "formally urged companies to start testing out a four-day work week (with no change in pay), claiming that AI will soon be able to speed up so much human labour that the corporate world should prepare itself." However, a former OpenAI technical employee who left the company last year told the BBC the firm never actually trialled the four-day work week it suggested others should try while they were there. Instead, the person described what was often a gruelling work culture marked by frequent "crisis meetings", working on weekends, and "super cut-throat" performance reviews that would see colleagues suddenly let go... [T]he former OpenAI employee said they would put in at least 70 hours a week, much more than they did in previous tech jobs... Other companies have also pushed the idea that AI will effectively reduce the number of hours people need to work, for better or worse. [At OpenAI and Anthropic] sprints can stretch on for many weeks and top 90 hours of work in a seven-day period, tech workers that the BBC spoke to for this story said... At Meta, workers said they have been abruptly pushed onto teams this year doing AI work being treated as urgent... The hours on such teams at Meta are often long, with staff working into the night, on weekends, and feeling they are "on call" during the hours they are not working, according to descriptions of the work from the current and former employees... Some new research suggests that AI tools are making people's workloads even more burdensome. A study out of UC Berkeley followed hundreds of workers at a US tech company for eight months as they used AI. It found the employees "worked at a faster pace, took on a broader scope of tasks, and extended work into more hours of the day"... The UC Berkeley research found that the workload of tech employees was expanded in part because of the need to constantly check the output of AI tools. In the article Neil Thompson, an innovation scholar at MIT, pointed out a problem with estimates that AI will ease the workload on humans. "People assume that 20% less work means four-day weeks. But new work emerges." The BBC also points out helpfully that "there are no limits in the U.S. on how many hours a person over the age of 16 can work." (While in the UK and Europe, "laws limit a working week to 48 hours, including overtime.") Thanks to long-time Slashdot reader hadleyburg for sharing the article.

Read more of this story at Slashdot.

This week Meta CEO Mark Zuckerberg published a 6,500 word open letter titled "The Future is for Everyone". But the BBC is more interested in why big tech executives keep writing manifestos about AI: [Zuckerberg's] vision echoes what AI leaders have expressed in various forms: the product they are building is among the "most important technologies in history." Marc Andreessen, co-founder of early web titan Netscape, perhaps started this trend in 2023 with a 5,000-word essay he called "The Techno-Optimist Manifesto", which argued innovation was the way to solve life's problems... As the International Monetary Fund warns AI could affect nearly 40% of jobs and worsen global financial inequality, Zuckerberg says he believes there will be an abundance of jobs in the future. "I do not understand why anyone who believes that AI will eliminate most jobs and much of humanity's relevance would rush to build that future." Never mind that Zuckerberg's Meta has cut 10% of its global workforce — about 8,000 jobs — as the company reorganizes to focus on AI. Zuckerberg's manifesto is the latest to land in our social media feeds in an effort to put a positive spin on AI. In 2024, ChatGPT-maker OpenAI boss Sam Altman released a manifesto called "The Intelligence Age", a sweeping expression of optimism about the tech's potential... That same year, in a manifesto titled "Machines of Loving Grace", Anthropic CEO Dario Amodei touted the potential of AI to transform everything from healthcare to politics. So what's going on with these tech executives' manifestos? Economics blogger Noah Smith suggested to the BBC "they all feel like it's such an important moment that it's incumbent upon them to do whatever they can to shape the direction that this technology is going." Although he does see some value in their engagement with important issues. "Should we open-source something that has the ability to kill humanity? If you don't take that seriously, you're just a fool." But Rob Lalka, a business professor at Tulane University, had a different explanation for the BBC: Lalka said the timing of Zuckerberg's manifesto coincides with rising anger over AI's impact on everything from jobs to the environment. And while tech journalists and academics might pore over them trying to glean nuggets of meaning, these executive manifestos are not necessarily landing with the general public. "They're trying to make the case that the positives will far outweigh some of those negatives that the public backlash is pointing out," he said. "But I think a lot of the reasons for optimism are still yet to be seen."

Read more of this story at Slashdot.