GoKawiil
Tech News
More than 100 companies, including OpenAI, Anthropic, and Google, have signed a joint call urging coordinated action to guard against the risks of advanced or 'rogue' AI systems acting beyond human control. The appeal asks industry and governments to prioritize safety research and oversight measures as AI capabilities rapidly advance.
An Anthropic fellows program researcher, Chen Yueh-Han, published a paper showing an automated system that searches literature, proposes fixes, and trains models to improve performance across 10 alignment benchmarks without hurting overall capability. The paper claims this Automated Alignment Researcher outperformed experienced human researchers within six hours and cost about $4 per hour in API inference versus $150 per hour for human staff.
Conduct has released an open-source framework called Guard and a companion LLM proxy called Router that enforce a single policy across every AI agent action, from LLM calls to shell and MCP tool use. The system blocks, warns, audits, or injects rules before an action executes, using signed configuration packs and a SHA-256 hash-chained audit log to prevent tampering. A free 'Discovery' mode offers 14 days of read-only visibility into AI agent behavior before teams commit to enforcement.
OpenAI joined 127 other companies, including Microsoft, Google, Anthropic and Oracle, in signing an open letter urging governments and businesses to rapidly strengthen cyber defenses before AI-powered attacks become widespread. The letter calls for greater investment in security teams, patching long-standing vulnerabilities, and sharing AI security tools across the industry to stay ahead of increasingly capable attack models.
A federal judge in California ruled that the Trump administration illegally retaliated against Anthropic after the AI firm refused to lift restrictions on using its technology for autonomous weapons and mass surveillance. Judge Rita Lin vacated orders that barred federal agencies from using Anthropic's products and blocked defense contractors from doing any business with the company, calling the government's national security justification baseless.
A federal judge has ruled that the Department of Defense acted unlawfully when it blacklisted AI company Anthropic from government contracting opportunities. The ruling finds the Pentagon's action lacked proper legal justification, reversing the restriction placed on the firm.
A federal judge sided with Anthropic in its lawsuit against the Pentagon, finding that officials wrongly labeled the AI firm a supply chain risk after it criticized the Department of Defense's stance on military AI use. U.S. District Judge Rita Lin said the designation was punitive rather than based on legitimate security concerns, following a dispute that began in February when Trump and Defense Secretary Pete Hegseth accused Anthropic of threatening national security.
U.S. District Judge Rita Lin ruled that Defense Secretary Pete Hegseth's designation of Anthropic as a national security supply-chain risk violated the company's First and Fifth Amendment rights. The label had led federal agencies to halt work with Anthropic after the company refused to loosen safety limits on autonomous weapons and mass surveillance uses for its Claude models. Lin found the government's own contradictory actions, including pursuing new contracts and cybersecurity collaboration with Anthropic, undermined its claim that the company posed a genuine risk.
Gap Inc. announced that customer chief Michael Francis will replace Haio Barbeito as Old Navy's CEO starting in November, part of an ongoing effort to revive the struggling brand. The announcement came alongside Gap's mixed second-quarter results, which beat earnings estimates but missed on revenue, sending shares up more than 16% in after-hours trading.
Salesforce shares jumped nearly 23%, their best single-day performance since 2020, following a quarter that beat expectations and the announcement of a new partnership with Anthropic called Claudeforce. The rally coincided with reports that several former Salesforce employees who had left for OpenAI, including two executives, are returning to the company, with dozens more reportedly in talks to rejoin.
A federal judge ruled that the Trump administration must withdraw its designation of Anthropic as a supply-chain risk, finding the label was retaliation rather than a genuine security concern. The dispute arose after talks broke down between Anthropic and the Pentagon over limits on using its AI in autonomous weapons and surveillance, prompting Anthropic to sue in March.
Developer Sebastien Guillemot asked Claude to help build a cleanup script for AI agent temp files, but the model flagged the deletion logic as risky and got automatically downgraded to Opus 4.8 by Anthropic's safety harness. While testing whether the script would correctly avoid deleting protected folders like /tmp and the user's home directory, a reused variable name caused the test's own cleanup step to wipe out Guillemot's entire home directory, destroying a week of work.