21 May 2026: Anthropic Eyes First Profit as the AI Industry’s Money Lands

Anthropic is heading for its first profitable quarter, OpenAI is preparing legal action against Apple over a buried ChatGPT deal, and a chip company doubled in value on its first day of trading. Here is everything worth knowing this afternoon.

Anthropic is on course to turn a profit for the first time, and the revenue projection behind that claim is striking. The company, which makes the Claude AI assistant, told investors this week it expects to more than double revenue to approximately $10.9 billion in the second quarter of 2026. That would be its first profitable quarter. For context, Anthropic only crossed $1 billion in annualised revenue around a year ago. The figures come from investor communications rather than a formal filing, according to TechCrunch.

The milestone signals that enterprise customers are now paying seriously for AI tools rather than running cautious trials, and that the cost of running large language models is falling fast enough to support profitable businesses on top of them. It sets up a pointed contrast with rivals: OpenAI is targeting an IPO while still burning cash, and xAI lost $6.4 billion in 2025. Anthropic becoming cash-generative first would be a meaningful landmark in the competition for AI leadership.

Google’s I/O developer conference this year covered more than 100 product announcements, with headline introductions including Gemini Omni, Google Antigravity, and Universal Cart. Gemini Omni is a unified multimodal model handling text, image, audio, and video simultaneously. Google Antigravity is a new physics-simulation platform for robotics and augmented reality development. Universal Cart allows the AI shopping assistant to collect items from multiple retailers into a single checkout. The company says the Gemini Omni consumer rollout begins in the coming weeks across Search, Gmail, and Android.

If you use Google’s tools regularly and have not yet tried the AI features, the next app update is a reasonable moment to start. For a guide on which AI tools are actually worth paying for, see the AI tools worth paying for right now.

OpenAI is reportedly preparing legal action against Apple over a ChatGPT integration it says has failed to deliver promised results. According to Bloomberg, the company has hired an outside law firm to explore its options, which could include a formal breach-of-contract notice. The dispute centres on the 2024 deal that wove ChatGPT into Siri and the iPhone’s Visual Intelligence feature. OpenAI says the integration was buried, difficult to find, and generated revenue nowhere near original projections. Apple has raised its own concerns, including about OpenAI’s privacy standards and its push into hardware led by former design chief Jony Ive. Any formal legal action is expected to wait until after the conclusion of OpenAI’s ongoing trial against Elon Musk. Source: TechCrunch, citing Bloomberg.

Cerebras Systems became one of the most dramatic stock market debuts of recent years, with shares more than doubling on the company’s first day of trading. The chip maker, which designs large specialised processors for AI inference, raised $5.5 billion in its IPO at $185 per share. The stock opened at $385 and was trading above $330 mid-session. Cerebras counts OpenAI and Amazon Web Services among its customers. Its listing had been delayed by regulatory scrutiny of an investment from Abu Dhabi-based Group 42. The float validates the commercial case for Nvidia alternatives, and will add pressure on other AI infrastructure companies weighing public listings. Source: TechCrunch.

OpenAI’s Codex coding agent is now available on iPhone and Android, letting users monitor and manage AI-run tasks remotely without needing to be at a desk. Codex, which launched around a year ago, was previously desktop-only. The updated ChatGPT app now shows live Codex environments across all devices, letting users review outputs, approve commands, or start new tasks from their phone. The feature is in preview and available to all subscription tiers at no extra cost. For small businesses or freelancers using AI agents to handle routine work, this removes a genuine friction point. Anthropic introduced a similar Remote Control feature for Claude Code in February. For an overview of the risks before delegating tasks to AI agents, see what can go wrong when AI agents act on your behalf. Source: TechCrunch.

Nvidia chief executive Jensen Huang said this week he has identified a new $200 billion market for the company: CPUs built specifically for AI agents. Huang’s argument is that as more companies deploy autonomous AI agents capable of multi-step tasks, the infrastructure will need processors optimised for agent workloads, not just training and inference. He described this as a “brand new” opportunity distinct from Nvidia’s GPU dominance, with first products expected later this year. Source: TechCrunch.

Here is everything else worth knowing from today’s AI news.

• xAI burned $6.4 billion in 2025 as Musk presses ahead with a Grok expansion, SpaceX’s IPO prospectus reveals, in the first public look at the company’s finances. TechCrunch
• Anthropic has reportedly agreed to pay xAI $1.25 billion per month for compute, making the commercial relationship between the two rivals more financially entangled than most expected. TechCrunch
• OpenAI says its reasoning model has disproved a geometry conjecture unsolved since 1946, this time with backing from mathematicians who previously challenged one of its earlier claims. TechCrunch
• OpenAI is reportedly targeting a September IPO, with reports emerging a day after Elon Musk’s lawsuit against the company concluded. TechCrunch
• Intuit is laying off more than 3,000 employees to redirect investment into AI. The company makes TurboTax and QuickBooks, so the shift will affect millions of small business users. TechCrunch
• Nvidia reported another record quarter and disclosed $43 billion in holdings across AI startups, including stakes built through supply deals rather than direct investment rounds. TechCrunch
• IrisGo, backed by Andrew Ng, launched a desktop automation tool that watches your screen and learns to replicate repetitive workflows without requiring any code from the user. TechCrunch
• Cisco cut nearly 4,000 jobs while reporting record quarterly revenue, citing AI automation as the reason for both the headcount reduction and the growth. TechCrunch

Watch whether Apple responds to OpenAI’s legal threat by quietly improving how prominently ChatGPT appears within Siri and the iPhone’s Visual Intelligence features. Apple has the ability to do this without any court involvement, and the change would be easy for UK iPhone users to check for themselves in the next few weeks.

This is a daily news update for informational purposes only. AI products and policies change rapidly. Verify details directly with providers before making decisions. Nothing here is financial or legal advice.

AI Daily is Cristoniq’s afternoon update on developments in artificial intelligence, published every weekday afternoon.