Google's bundling of Search access with AI tool adoption in schools has triggered a political firestorm, exposing the market-power risks of vertically integrated AI education products. ⚖️
Happy Wednesday. AMD just posted its biggest quarter ever — $13.4B in revenue with data center sales nearly doubling — and guided to $17B next quarter, a powerful signal that AI infrastructure spending remains in full acceleration. Meanwhile, the day's other big themes: a flurry of AI security moves (CrowdStrike's kill switch, its twin AI-startup acquisitions), Salesforce snapping up Cresta for $2.7B, and India's TRAI quietly making a consequential move by mandating AI agents in telecom customer support.
AI is moving into production across a surprisingly wide range of verticals today — from Cisco's machine-speed cyber defense and Heimstaden's property management transformation, to LiveX's avatar kiosks in retail — while Palantir's earnings reveal a structural milestone: US commercial AI revenue has overtaken government revenue for the first time.
Enterprise AI platform buildout is accelerating on two fronts today: CrowdStrike is embedding agentic AI safety controls directly into its security platform (including an 'AI kill switch'), while Snowflake and IBM are racing to make agentic workloads governable and production-ready for large organizations.
Anthropic is quietly working on a novel authentication layer for AI model identity, while OpenAI continues to showcase enterprise model deployments in engineering and real estate — both pointing to a maturing ecosystem where trust and provenance of AI outputs are becoming as important as capability.
AMD's historic quarter — $13.4B in revenue with data center sales nearly doubling and a $17B guide — is the clearest proof yet that AI infrastructure demand has not peaked; meanwhile, creative power solutions from jet engines to fuel cells underscore the grid bottleneck that remains the sector's defining constraint.
Today's deal flow is defined by consolidation in AI security and aggressive bets on generative media infrastructure, with Salesforce's $2.7B Cresta acquisition and CrowdStrike's twin AI-safety buyouts signaling that large incumbents are acquiring their way into agentic AI capabilities rather than building from scratch.
AI is hitting regulatory and legal friction on multiple fronts simultaneously: an EU court adviser is warning that Google's AI chatbot may violate copyright law, AI-generated political ads are flooding US midterm campaigns, and Delhi Police's roaming facial recognition vans are prompting civil liberties questions in India.
India is seeing meaningful AI policy and market developments simultaneously: TRAI has formally authorized AI agents in telecom customer support (a regulatory unlock with millions of users in scope), two indigenous LLM startups launched new models, and Jio Platforms posted strong Q2 growth that underpins the country's digital-AI infrastructure ambitions.
Hiding inside Palantir's skyrocketing quarter is a foreboding signal — The moment US commercial AI revenue at Palantir overtook government revenue isn't just a Palantir story — it's the first clean data point that enterprise AI adoption has crossed a structural inflection point. Senior strategists need to understand what's driving that commercial surge and whether it signals a broader shift in where AI ROI is actually landing. Read →