NewsletterIssue 3·29 September 2026

AI³ Newsletter, Issue 3

This Blog in Five

OpenAI called for mandatory safety rules for the most powerful AI systems.
The EU brought ChatGPT, Reddit and Roblox under stronger Digital Services Act oversight.
OpenAI and Anthropic both released new, cheaper frontier models.
Meta went all-in on its Muse AI agent and new consumer AI devices.
SoftBank raised $11.1 billion in bonds to help fund its OpenAI investment.

Policy & Regulation

OpenAI Calls for Tougher Frontier AI Rules

OpenAI is urging governments to introduce mandatory safety requirements for the most powerful AI systems, including independent testing, cybersecurity standards and incident reporting. The company argues that AI is increasingly helping accelerate AI research itself, making early regulation more urgent. It also warns that rules should target frontier developers without creating barriers that permanently protect today's largest AI companies.

EU Brings ChatGPT, Reddit and Roblox Under Stronger DSA Oversight

The European Commission has designated ChatGPT as a Very Large Online Search Engine under the Digital Services Act, while Reddit and Roblox have been classified as Very Large Online Platforms. The move brings tougher obligations around systemic risks, transparency and user protection. For ChatGPT, it also signals that Brussels increasingly sees AI assistants as major gateways to online information.

ENISA Begins Testing Anthropic's Cyber AI

The EU Agency for Cybersecurity has begun testing Anthropic's Mythos 5 model after gaining access to the restricted cybersecurity system. The development highlights Europe's growing need to evaluate frontier AI tools for cyber defence, while also exposing its dependence on advanced models developed and controlled outside the EU.

Bill Gates Warns Governments Are Falling Behind on AI

Bill Gates has warned that governments are not prepared for the speed of AI-driven change. He highlighted risks including job displacement, cyber threats, addictive AI companions and widening inequality, arguing that policymakers need stronger expertise and coordinated oversight before these effects become harder to manage.

OpenAI and Anthropic Release New, Cheaper Models

Both companies rolled out new frontier models pitched on cost efficiency, with OpenAI introducing GPT-6 Sol and GPT-6 Luna, and Anthropic launching Claude Opus 5.5. The simultaneous pricing push signals an intensifying price war as enterprises scrutinise AI spending. It also follows the CEOs' recent public calls for a more measured pace of development, suggesting competition remains the dominant driver.

Meta Goes All-In on Muse Agent and Consumer AI Devices

At its annual Connect event, Meta announced a major expansion of its AI agent Muse, including integration with the new Muse Charm device and Meta's AI glasses. The move puts Meta ahead of rivals like OpenAI and Google in consumer AI hardware, though the commercial strategy remains unproven. Tensions have already emerged, with Amazon blocking the Muse app from its site, a possible sign of coming platform conflicts over AI product distribution.

SoftBank Raises $11.1 Billion in Bonds to Fund OpenAI Investment

SoftBank shares jumped more than 7% after the Japanese tech investor issued $11.1 billion in bonds to help finance its bet on OpenAI. The strong market reaction shows continued investor appetite for AI exposure at scale. It also illustrates how major AI investments are now being funded through debt markets, a shift worth watching as capital commitments to AI keep growing.