AI news · Tuesday, June 2, 2026

Uber capped AI spending at $1,500 a head after blowing a year's budget in four months

The AI gold rush is hitting a massive wall of reality as tech giants pivot from experimental hype to proving real financial returns. Uber is slashing its AI budget, capping employee spending at $1,500 a month after burning through an entire year’s budget in just four months, signaling that the era of unlimited AI experimentation is officially over. Meanwhile, Microsoft is going all-in on controlling the chaos by launching 'ASSERT' to force AI agents to behave under strict policy guardrails and the 'Agent Control Specification' to standardize how these agents operate across different software.

Even with these guardrails, companies are scrambling to make AI actually useful for regular people, with Microsoft launching 'Scout' to act as an always-on assistant in Microsoft 365, mirroring the agentic shift seen in OpenAI’s latest Codex updates, which now boast over 5 million weekly users. The competition for enterprise supremacy is heating up as Google pushes its own agent, Gemini Spark, into the workspace. The stakes are enormous because developers are now racing to build hardware designed specifically for these agents, with Microsoft debuting its 'Project Solara' platform to power everything from wearable badges to desktop hubs.

Financial heavyweights are taking massive positions as Berkshire Hathaway bets $10 billion on Alphabet, signaling a major move into the AI infrastructure space. The market is reacting wildly to these shifts, with Hewlett-Packard Enterprise stock surging 30% after a massive earnings beat driven by cloud and AI revenue. However, the workforce impact remains volatile, as tech leaders like Mo Gawdat warn that 30% of certain job sectors could vanish by 2028, forcing workers to double down on 'human-centric' skills that AI currently struggles to replicate.

Cybersecurity is becoming the biggest battleground, with Anthropic expanding its 'Project Glasswing' initiative to over 150 organizations across 15 countries to hunt for critical software vulnerabilities using its most powerful model, Claude Mythos. Google is also launching fake call detection on Android to combat AI-powered deepfake impersonation scams that cost Americans nearly $900 million last year. While some Hollywood legends like Martin Scorsese are dipping their toes into AI for storyboarding, the broader industry is facing a massive test as indie hits like 'The Amazing Digital Circus' prove that fan-driven, independent animation can absolutely dominate the global box office.

The quick hits

Uber capped employee AI spending at $1,500 per month after burning through their entire annual budget in just four months—showing companies are finally demanding real returns on investment.
Microsoft launched Scout, an always-on assistant for Microsoft 365—this acts like a permanent digital coworker that can manage your calendar, draft emails, and handle routine work tasks.
Google is rolling out fake call detection to Android—this uses real-time signals to alert you if a scammer is using AI deepfakes to impersonate your family or contacts.
Anthropic is expanding Project Glasswing to 150 organizations—they are using AI to find and fix thousands of critical security flaws in the world's most important software infrastructure.

Uber capped AI spending at $1,500 a head after blowing a year's budget in four months

The quick hits

Sources