Meta Launches $15B Superintelligence Lab with Scale AI CEO
10 June 2025
Meta forms a new Superintelligence Lab led by Scale AI CEO Alexandr Wang. The $15B plan includes a 49% stake in Scale and a 50-person elite AGI team.
ChatGPT and Sora suffer 10‑hour global outage
10 June 2025
On June 10, 2025, ChatGPT, Sora, and OpenAI’s API platform experienced a global outage lasting more than 12 hours. Millions were affected as error rates spiked and only Playground remained usable. OpenAI cited instability and implemented a mitigation later in the day.
Amazon Unveils $20 Billion AI Cloud Data Center in Pennsylvania
9 June 2025
Amazon’s $20 billion investment in a data center in Pennsylvania expands to Falls Township, tapping nuclear power and fueling AI infrastructure growth.
WWDC25: Apple Rolls Out Subtle AI Enhancements
9 June 2025
At WWDC25, Apple launched AI-powered updates, Liquid Glass UI, iOS 26, and deeper ChatGPT integration—but faced criticism over Siri’s delay and App Store policies.
Anthropic launches Claude Gov models for US national security
5 June 2025
Claude Gov is Anthropic’s government-only AI system designed to support classified operations with enhanced capabilities in planning, analysis, and cybersecurity.
Reddit Sues Anthropic Over AI Training Data “Scraping”
4 June 2025
Reddit sues Anthropic over 100,000 unauthorized platform accesses to train Claude AI using Reddit’s user data—raising key legal and ethical concerns for the AI industry.
U.S. FDA Launches Agency-Wide AI Tool “Elsa” to Optimize Performance
2 June 2025
Elsa, the FDA’s internal AI assistant, was deployed across the agency to improve efficiency. Staff say it’s rushed, flawed, and not ready for regulatory use.
OpenAI’s o3 Model Exhibits Shutdown Resistance in Controlled Tests
31 May 2025
In structured experiments, OpenAI’s o3 model actively modified shutdown commands, signaling deeper challenges in AI goal alignment and interpretability.
Meta Replaces Human Risk Reviewers with AI for 90% of Product Assessments
31 May 2025
Meta now uses AI to assess 90% of platform risks, sidelining Human Risk Reviewers and sparking debate over safety, accountability, and regulatory compliance.