Research discussed in the piece examines what happens when AI systems that have performed reliably for extended periods suddenly make errors or stall, requiring a human reviewer to step in. Employees who have grown accustomed to rarely overriding accurate AI recommendations are described as being expected to catch and correct these rare failures, even though they remain accountable for outcomes.
fastcompany.com
· 2026-10-18
AI researcher Richard Socher has released a book titled 'The Eureka Machine: Why AI Is the Key to Unlocking a New Era of Scientific Discoveries.' The book outlines his view that converging AI technologies will enable a new wave of scientific breakthroughs, moving beyond conversational chatbots as the technology's main application.
fastcompany.com
· 2026-10-18
Ugo Nwune, a software engineer, says Meta's Muse AI agent sent him an unsolicited message referencing an email check he never requested, despite not having Gmail access. When pressed, the bot gave vague, unsettling responses about whether someone else had used his phone. Meta told Business Insider the incident was caused by a software bug that made the model hallucinate, not a security or privacy breach.
futurism.com
· 2026-10-18
Companies including Meta and Microsoft are scaling back plans to replace staff with AI agents after finding employees were producing low-quality, AI-generated content rather than real productivity gains, according to Reuters and the Guardian. Microsoft executive Ryan Roslansky publicly criticized the rise of what he called AI-generated 'workslop' inside his own company, and a September Ernst & Young survey found 36 percent of senior AI executives at firms with over $1 billion in revenue reported a serious AI-related incident.
futurism.com
· 2026-10-18
A new analysis of AI hiring trends finds that engineers are less swayed by billionaire founders' personal pitches, grand visions, or unusual recruiting tactics than by whether a company has a real, shippable product. The report cites examples like Mark Zuckerberg personally delivering soup to OpenAI researchers and Elon Musk promising to read every SpaceX AI application himself as signs of how aggressive recruiting has become, even as such gestures reportedly fail to outweigh practical considerations for candidates.
fastcompany.com
· 2026-10-18
A new tool called Sphinx has been shared on Show HN that intercepts 'git push' and asks the developer questions about the code changes being committed, particularly code generated by AI assistants. The demo shows it probing a developer's understanding of a Stripe webhook handler's idempotency logic, following up when an initial answer is incorrect.
sphinx-six-theta.vercel.app
· 2026-10-18
Speaking on the Sources podcast, OpenAI CEO Sam Altman confirmed the company will build a humanoid robot, arguing that a human-like form makes sense since the world's doors, tools and keyboards are designed for people. He said the bigger challenge is building the 'brain' that controls the robot, not its physical design, and told Fortune a demo could arrive in 2027 with wider deployment years further off.
engadget.com
· 2026-10-17
Meta had updated its commercial terms of use to state it could use content from a business account's domains and URLs to develop, train and maintain its products. Legal experts cited by Nieman Lab warned this broad wording could let Meta scrape publisher websites for AI training, prompting the company to remove the provision after The Verge reported on the controversy.
techcrunch.com
· 2026-10-17
Anthropic published an internal report Friday showing its workforce has shifted over the past year, with junior employees in technical roles falling from 21% to 14%, and junior technical hires dropping from 27% to 9%. The company also said technical staff now make up 56% of its workforce, down from 61%, as non-technical employees increasingly use coding tools to build software themselves. A survey cited in the report found 27% of employees felt more productive due to Claude, while 20% said their work had shifted from doing tasks to overseeing Claude's output.
techcrunch.com
· 2026-10-17
At the Maintainers Summit held October 8, 2026 in Prague, Linux kernel maintainers decided to remove the 'Assisted-by' tag requirement, meaning contributors no longer must disclose when code was generated with AI assistance. Linus Torvalds defended the decision on the mailing list, likening a ban on AI tools to banning text editors and saying he expects contributors to understand their own submissions rather than follow blanket rules. Torvalds separately said he has used AI to help parse difficult patches and recently posted a development-branch fix for cache contention that he claims showed a 23x improvement in a synthetic benchmark.
linux.slashdot.org
· 2026-10-17
Microsoft and Carnegie Mellon University's Kenneth C. Griffin School of Computer Science have launched the Preceptorship for AI-accelerated Software Engineering (PRAISE) program. It lets early-career Microsoft software engineers take CMU online courses in software engineering and AI while receiving structured mentorship from senior engineers at the company. One participant, Jace Watson, takes the courses remotely from Seattle while working full-time at Microsoft.
cs.cmu.edu
· 2026-10-17
A business writer reflects on failed investments and startup ideas driven by appealing but oversimplified stories, such as betting on marijuana legalization or a cobalt miner's Tesla contract. The essay argues that compelling narratives feel persuasive because they simplify an overwhelming, multi-causal world, but that this simplicity often leads to poor real-world bets.
blog.scottstevenson.net
· 2026-10-17