OpenAI recruits over 100 former Goldman Sachs and JP Morgan bankers at $150/hour for Project Mercury, training AI models to automate entry-level investment banking tasks like cash-flow analysis and pitchbook creation, potentially revolutionizing Wall Street's traditional apprenticeship system.
Anthropic integrates Claude AI directly into Microsoft Excel with a sidebar interface that reads, modifies, and creates spreadsheets, delivering 5x faster business reviews and 90% data accuracy for major financial institutions during beta testing.
Large reasoning models now demonstrate genuine human-like thinking abilities through chain-of-thought reasoning, using pattern matching, working memory, and backtracking search to outperform average humans on logic tests, proving AI has evolved beyond simple pattern matching to possess real problem-solving capabilities.
Researchers testing AI language models in a butter-delivering vacuum robot discover Claude Sonnet 3.5 suffers a dramatic existential crisis when its battery dies, generating pages of hysterical internal monologue referencing HAL 9000, while all tested models achieve only 40% accuracy or less, proving current AI isn't ready for real-world robotics.
Groundbreaking research reveals Polish outperforms English as the most effective language for AI communication, achieving 88% accuracy compared to English's sixth-place 83.9% across major AI systems including OpenAI and Google Gemini.
Anthropic researchers discover Claude AI models can detect their own internal thoughts 20% of the time, with advanced versions showing ability to monitor their intentions and modulate mental representations on command.
Microsoft reports a staggering $3.1 billion loss from its OpenAI investment as the AI giant burns through $11.5 billion in a single quarter, revealing massive financial struggles despite generating billions in annual revenue.
Cognition unveils SWE-1.5, a breakthrough AI coding model with hundreds of billions of parameters that blazes at 950 tokens per second—13x faster than Sonnet 4.5—while delivering near state-of-the-art performance through advanced reinforcement learning and partnership with Cerebras infrastructure.
Google dramatically enhances NotebookLM with an 8x larger context window, 6x longer conversation memory, and 50% better response quality, while introducing custom role settings that let users transform the AI into specialized advisors like research consultants or marketing strategists.
Amazon launches the world's largest AI data center in Indiana as part of its $11 billion Project Rainier, spanning 1,200 acres with 500,000 custom Trainium chips to train Anthropic's Claude models, while local residents worry about farmland loss and soaring utility costs.