Daily Recap

Friday, 11 September 2026

Tech news today centers on a widening struggle to govern powerful digital platforms and AI systems as they become more capable and more contested. Companies and governments are weighing how fast to move, how to police misuse, and how much accountability to impose when safety claims fall short.

Sam Altman says OpenAI is open to slowing frontier AI development

OpenAI is open to slowing development of frontier AI systems, with CEO Sam Altman discussing the possibility internally as concern grows over the risks posed by increasingly capable models. The shift signals that safety pressures are now shaping strategy at the highest level of one of the industry’s most influential AI companies.

Texas judge rules TikTok misled parents about child-safety protections

A Texas judge rules that TikTok violates state consumer protection law by misleading parents and users about safeguards for minors on the platform. The summary judgment finding under the Texas Deceptive Trade Practices Act marks a significant legal setback for TikTok in a case focused on child-safety claims and platform accountability.

Anthropic says it disrupted AI misuse tied to state surveillance by China and Iran

Anthropic says it disrupted several state-linked surveillance operations tied to China and Iran that used its AI models between January and July. The company says the activity targeted ethnic minorities and dissidents, underscoring how governments and aligned actors are adopting generative AI tools for monitoring, repression, and influence operations.

Anthropic says Yemen weapons cell used its Claude AI to aid guided missile development

Anthropic also says it shut down misuse of its Claude model by a cell in northern Yemen that used the system to support weapons development. The company says the group ran three separate programs and applied the AI in work related to guided airborne weapons, highlighting the security risks around access to advanced models.

UK Cabinet Office rejects emergency “AI kill switch” proposal

The UK government rejects an emergency AI “kill switch” that would let authorities block access to dangerous models, saying Britain cannot simply turn AI off. The decision shows officials favoring other forms of oversight over a hard-stop mechanism, even as debate intensifies about how to respond to high-risk AI failures.

Tencent-Backed Chipmaker Enflame Soars in Shanghai Debut After $911 Million IPO

Chinese chipmaker Enflame surges in its Shanghai STAR Market debut after raising $911 million in its IPO, with the stock jumping well above its offer price on the first day of trading. The strong opening highlights intense investor appetite for domestic semiconductor players tied to China’s push for homegrown AI and computing infrastructure.

Rockstar North worker tells tribunal he believes staff were put on a “watch list”

At an employment tribunal, a dismissed Rockstar North worker says he believes staff were placed on a “watch list” after signing a petition over remote working. The allegation emerges in a broader dispute involving claimed trade union activity and possible unlawful dismissal at the studio behind Grand Theft Auto.

Cyble appoints cybersecurity veteran Steve Ingram as US Executive Vice President

Cyble names cybersecurity veteran Steve Ingram as Executive Vice President for the United States, giving him responsibility for leading the company’s expansion in the market. The appointment moves him from an advisory role into operating leadership as the firm looks to strengthen its U.S. presence.

Recap for Friday, 11 September 2026

Photos from today's Tech coverage
11 Sep Top Stories

Start typing to search stories...