AI Trends — 27 May 2026
Your daily digest of the most important developments in artificial intelligence.
🔬 OpenAI AI Solves 80-Year-Old Math Problem
In what researchers are calling a landmark moment, OpenAI announced that one of its general-purpose reasoning models autonomously cracked a famous geometry problem that had stumped mathematicians for more than eight decades. The implications are significant: an AI system capable of original mathematical discovery could eventually accelerate breakthroughs across science, engineering, and medicine. This marks one of the first clear demonstrations of AI performing genuinely novel academic research rather than reciting trained knowledge.
💰 Anthropic on Track for First Profitable Quarter
Anthropic's explosive growth has placed the company on the cusp of a historic milestone — its first profitable quarter. Revenue is set to more than double to $10.9 billion in Q2 2026, with an estimated $559 million operating profit arriving two full years ahead of internal projections. Adding to that momentum, Anthropic also expanded its compute partnership with SpaceX, agreeing to spend roughly $1.25 billion per month through 2029 for access to the company's Colossus supercomputing infrastructure.
⛪ Pope Leo XIV Calls to "Disarm" AI
In a sweeping 42,300-word encyclical titled Magnifica humanitas, Pope Leo XIV issued a major warning about the dangers of artificial intelligence, urging global leaders to "disarm" AI before it enables domination, exclusion, and harm. The document calls for AI to serve humanity rather than concentrate power in the hands of a few. Anthropic co-founder Chris Olah publicly commented on the encyclical, underscoring how seriously the AI safety community is taking its message.
🍎 Apple Prepares to Open AI Ecosystem
Apple is preparing a significant platform shift that would allow users to select third-party AI providers — including Google and Anthropic — to power Apple Intelligence features across iOS 27, iPadOS 27, and macOS 27. The move signals a major strategic pivot, acknowledging that users want choice in their AI assistant rather than a single first-party option.
🤖 Anthropic Unveils "Dreaming" Agent Technique
Anthropic researchers unveiled a new AI agent technique called "dreaming" — designed to help autonomous systems review their prior behavior, identify patterns, and improve future performance between sessions. The technique is part of broader efforts to build more capable managed agents for long-running workflows in domains like coding, finance, and law.
🏛️ Trump Signs AI Executive Order with Big Tech CEOs
President Trump summoned top tech leaders to Washington for the signing of a new AI executive order. The guest list included Alphabet's Sundar Pichai, Apple's Tim Cook, OpenAI's Sam Altman, and Anthropic's Dario Amodei. The order is expected to establish a voluntary framework requiring AI labs to share new models with the government 90 days before public release — an approach seen as a middle ground between light-touch regulation and formal oversight.
🌍 UK and Australia Sign AI Security Pact
The United Kingdom and Australia formalized a new security agreement focused on fast-moving AI risks. The pact aims to align both nations on shared standards for evaluating AI systems, particularly in the context of national security and critical infrastructure.
📊 By the Numbers
- $10.9B — Anthropic's projected Q2 2026 revenue
- $19.8B — JPMorgan Chase's 2026 technology budget, with 2,000 staff dedicated to AI
- 3.5× — How much more AI "intelligence" frontier companies use per employee vs. typical firms
- 100× — Energy reduction achieved by a new AI efficiency algorithm without sacrificing accuracy
- 42,300 — Words in Pope Leo's AI encyclical
🔭 Looking Ahead
The coming days will be shaped by how the White House's new voluntary AI framework lands with international regulators — particularly in the EU, where mandatory rules are already in effect. Watch also for reactions to OpenAI's math breakthrough: if replicated, it could reset expectations for what "general intelligence" means in practice.
Sources: Time · Al Jazeera · Axios · CNBC · InfoWorld · imfounder.com · GOV.UK