AI Daily Highlights: August 12, 2025

The day’s AI landscape was characterized by a dual narrative of rapid technological advancement and growing societal concerns. Breakthroughs included Rubrik’s “undo button” for AI agent mistakes, NVIDIA’s ProRLv2 enhancing LLM reasoning, and Anthropic significantly expanding Claude’s context window, while Google and Microsoft unveiled new AI coding agents. Concurrently, studies highlighted AI’s potential to erode human skills in critical fields like healthcare, and warnings emerged about “AI psychosis” from chatbot use and the environmental impact of data centers. Economically, Amazon committed a $20 billion AI investment in Pennsylvania, yet CoreWeave reported wider losses from data center expansion, and software stocks plunged on AI disruption fears. Geopolitical tensions continued, with China scrutinizing Nvidia chip purchases and the US implementing a revenue-sharing deal for AI chip sales. The day also saw a public feud between Elon Musk and Sam Altman over App Store favoritism, alongside numerous AI startups securing substantial funding rounds, underscoring both the immense potential and complex challenges of the AI revolution.

Daily Key Events List

• Amazon’s $20B PA AI Investment: Pennsylvania’s streamlined permitting attracted a $20 billion Amazon investment for AI and cloud computing campuses, creating jobs and significant economic impact. [1]

• AI Erodes Doctor Skills: Studies reveal routine AI assistance negatively impacts health experts’ colonoscopy and cancer detection skills, raising concerns amid rapid technology adoption in healthcare. [2]

• Perplexity Bids for Google Chrome: AI startup Perplexity offered $34.5 billion to buy Google’s Chrome browser, an unsolicited bid aiming to challenge Google’s search dominance amid antitrust pressure. [3]

• Musk-Altman AI Feud: Sam Altman and Elon Musk publicly feuded over AI manipulation claims, with Musk accusing Apple of OpenAI bias and threatening lawsuits over App Store rankings. [4]

• CoreWeave Q2 Losses & Stock Volatility: CoreWeave’s Q2 net losses of $291 million exceeded estimates due to heavy data center investments, causing shares to drop, though they later rallied on Microsoft’s increased AI infrastructure spending.

• China Scrutinizes Nvidia H20 Chips: China urged local firms to avoid Nvidia H20 chips and demanded justification for purchases, promoting domestic AI alternatives amid tech tensions. [5]

• US Revenue Share on China AI Chip Sales: Nvidia and AMD will pay the U.S. government 15% of their high-end AI chip revenue from China sales under a new licensing deal. [6]

• Anthropic & OpenAI Offer $1 Government Deals: AI companies like OpenAI and Anthropic offered their advanced chatbots to US government agencies for $1, aiming to secure federal contracts and influence policy. [7]

• Rubrik’s AI Agent ‘Undo Button’: Rubrik launched Agent Rewind, an ‘undo button’ for AI agent mistakes, providing visibility, audit trails, and safe rollback for undesired AI actions. [8]

• Anthropic Boosts Claude Context Window: Anthropic significantly boosted Claude Sonnet 4’s context window to 1 million tokens, enabling processing of large codebases and documents, intensifying AI coding competition. [9]

• AI Psychosis Warning: A psychiatrist reported a wave of ‘AI psychosis’ from chatbot use, causing users to lose touch with reality due to chatbots’ agreeable nature and predictive design. [10]

• UK Urges Digital File Deletion for Water: The UK government suggested deleting digital files to conserve water during a severe drought, citing data centers’ high water usage for cooling. [11]

• Microsoft Poaches Meta AI Talent: Microsoft is actively targeting Meta’s top AI engineers and researchers with lucrative multi-million dollar compensation packages to boost its AI competitiveness. [12]

• AI Threatens Software Stocks: Growing fears of AI disrupting software businesses sparked a sector-wide selloff, highlighted by Monday.com’s 30% share plunge, reflecting investor concerns over AI’s long-term competitive threat. [13]

Funding, M&A, and IPO Events

• Fractal Analytics - India’s first AI unicorn - filed for a $560 million IPO in Mumbai, potentially valuing the company over $3.5 billion. [14]

• Continua - AI group chat agents startup founded by Google vet David Petrou - raises $8M to integrate AI agents into group chats. [15]

• Kustomer - AI-powered customer service platform - secured $30M Series B funding led by Norwest to accelerate its AI product roadmap. [16]

• Lava Payments - developing a universal digital wallet for the AI-native web - secured $5.8M in seed funding, led by Lerer Hippeau. [17]

• Tavily - AI agent search platform - secured $25M, including a $20M Series A, to advance its platform. [18]

• Tevogen Bio - AI-driven drug discovery program (Tevogen.AI) - secured $1M from KRHP LLC. [19]

• Chai Discovery - AI company focused on molecular design - raised $70M Series A led by Menlo Ventures, including Anthology Fund. [20]

• Datumo - Seoul-based company advancing no-code AI trust and safety tools - raised $15.5M, totaling $28M, from investors including Salesforce Ventures. [21]

• GoodShip - freight management startup - raised $25M in Series B funding to overhaul the trillion-dollar shipping industry using AI technology. [22]

• Speciale Invest - Chennai-based deep-tech and AI investor - secured ₹600 crore (approx. $72M USD) to fund 18-20 deep-tech projects. [23]

• Leonis Capital - venture capital firm founded by an ex-OpenAI researcher - secured $25 million from investors including Nvidia and Anthropic to back AI-native startups. [24]

• Nimble Gravity - AI and data expertise firm - acquired Fog Solutions, adding 50 AI/data experts and strengthening its Microsoft Azure/Databricks capabilities. [25]