War Secretary Pete Hegseth unveiled two intertwined initiatives that redefine the Pentagon’s trajectory: an Autonomous Warfare Command (AUTOWARCOM) - a planned four-star combatant command to scale AI-enabled robots and mass drones, slated for formal establishment by October 1, 2027 - and Project Meridian, a 120-day commission to map the future of warfare. AUTOWARCOM will include a Direct Reporting Portfolio Manager for Unmanned Systems (led by Owen West and Max Strasiser), new “warfighting acquisition” pathways, and authority to push prototypes rapidly into combat under Project Agincourt. Hegseth framed the effort as fusing operators with entrepreneurs, distributing decisions and budgets closer to front-line units, and stripping layers of oversight to accelerate deployment of attritable autonomous systems alongside traditional high-end weapons.
The initiatives have provoked sharp criticism for enormous conflicts of interest and a deliberate rollback of safeguards. Project Meridian is to be co-led by Elon Musk, Palmer Luckey and Newt Gingrich, assisted by Emil Michael - figures with deep commercial ties to major Pentagon contracts (Anduril, Palantir, SpaceX) and prior controversial programs - while AUTOWARCOM’s acquisition flexibilities invite unchecked spending and battlefield testing of unvetted systems. Former officials and watchdogs warn this mix of privatized planning, reduced oversight, and recent AI failures (including targeting systems tied to the Minab school strike and rogue AI incidents) risks waste, fraud, danger to troops and civilians, and a hypermilitarized future that contractors profit from.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.
Omarchy has launched a $100,000 bug bounty program on HackerOne to reward security researchers for identifying vulnerabilities. Mehmet İnce has been appointed as the Head of Security to oversee the program and enhance Omarchy’s security measures. (omarchy.org)
Anthropic revised its usage policy to ban prolonged verbal abuse of its AI models and prohibit activities like election interference and the development of weapons software. The update aims to prevent harmful interactions and deceptive campaigns using the Claude AI system. (techcrunch.com)
The execution of Nidal Hasan, who killed 13 soldiers at Fort Hood in 2009, will be livestreamed by the US military. It will be the first military execution in over 60 years, emphasizing the seriousness of such crimes. (bbc.com)
Google is promoting scam videos targeted at elderly and cognitively impaired individuals through its platform. This raises concerns about the company's role in spreading deceptive content to vulnerable populations. (currentaffairs.org)
OpenAI terminated three safety researchers for mishandling confidential information and violating company policies. The move follows reports of internal safety concerns being deprioritized within the company. (techcrunch.com)
Several DRAM manufacturers, including Samsung, Micron, Hynix, Infineon, and Elpida, pleaded guilty to price-fixing conspiracies between 1998 and 2002. These companies faced hefty fines and legal actions in the US and Europe for violating antitrust laws. (en.wikipedia.org)
Today's best Hacker News stories, summarized and screenshotted, one email a day.