Weekly AI Roundup: February 28, 2026
Published February 28, 2026
It was another consequential week in artificial intelligence. From record-setting funding rounds to government partnerships that challenge the industry's relationship with defense, here are the biggest AI stories from the week ending February 28, 2026.
OpenAI Closes Massive New Funding Round
OpenAI confirmed this week that it has closed a new funding round reportedly valued at $40 billion, making it the most valuable private technology company in the world. The round was led by SoftBank's Vision Fund, with participation from existing investors including Microsoft, Thrive Capital, and Andreessen Horowitz. CEO Sam Altman stated the funds will accelerate development of next-generation reasoning models and expand compute infrastructure. Critics argue the valuation reflects speculative hype rather than sustainable revenue, while supporters point to OpenAI's rapidly growing enterprise business as validation.
Anthropic Deepens Pentagon Partnership
Anthropic, the AI safety-focused startup behind the Claude family of models, announced an expanded partnership with the U.S. Department of Defense this week. The deal involves providing Claude models for intelligence analysis and logistics optimization — applications the company says fall within its responsible use guidelines. The announcement triggered debate within the AI safety community. Some researchers argue that working with defense agencies contradicts Anthropic's stated mission of building safe AI, while others contend that responsible engagement with government is preferable to leaving the field to less safety-conscious competitors.
Google Rolls Out Gemini 2.5 Pro
Google DeepMind released Gemini 2.5 Pro this week, featuring significant improvements in reasoning, code generation, and multimodal understanding. The model showed strong benchmark results in mathematical reasoning and scientific problem-solving. Google also announced deeper integration of Gemini into Workspace products, including real-time document summarization in Google Docs and automated data analysis in Sheets. The release intensifies the competition with OpenAI's GPT series and Anthropic's Claude.
EU AI Act Enforcement Begins
The first enforcement provisions of the European Union's AI Act took effect this week, with new transparency requirements for AI systems deployed in EU member states. Companies using AI for content recommendation, hiring decisions, or credit scoring must now disclose the use of AI to affected individuals and provide mechanisms for human review. Several major tech companies have announced compliance updates, though critics argue the regulations still lack sufficient teeth for enforcement.
AI-Generated Content Floods Social Media Ahead of Elections
Researchers at Stanford's Internet Observatory published a report documenting a significant surge in AI-generated political content across social media platforms in February. The report identified coordinated campaigns using synthetic images, voice clones, and AI-written posts targeting elections in multiple countries. Platform responses have been mixed — Meta announced new AI-content labeling requirements, while X (formerly Twitter) has been slower to implement detection measures. The report underscores the growing urgency of AI literacy for voters and media consumers.
What to Watch Next Week
Looking ahead: NVIDIA's GTC conference kicks off next week with expected announcements about next-generation AI chips. Apple is also rumored to preview new AI features for iOS 20 at its spring event. And the ongoing congressional hearings on AI regulation continue with testimony from industry leaders.
For daily AI coverage, visit our AI section. To understand how we curate these stories, read how KhanList curates news.