AI Is Getting Real: OpenAI Cuts Ties With Cursor, Anthropic Wins in Court, and AI-Powered Hacks Hit Companies

OpenAI Is Ending Its Cursor Partnership

OpenAI said it plans to terminate its agreement with Cursor, the AI coding company now owned by SpaceX. OpenAI cited a lack of confidence that SpaceX would use its technology within OpenAI's terms of service.

For developers, the bigger lesson is about dependency.

Modern coding assistants often rely on several model providers. A developer may see one product on the screen, but the underlying models can come from multiple companies. If one provider changes its licensing or partnership policy, the product can change overnight.

Businesses should therefore look beyond features when choosing AI development tools. Model access, licensing and supplier stability matter too.

Anthropic Wins a Major Legal Battle

A U.S. federal judge ruled that the Pentagon's decision to designate Anthropic as a supply-chain risk was unlawful.

The dispute grew out of Anthropic's refusal to remove certain restrictions on the use of Claude for autonomous weapons and mass surveillance.

The case matters beyond Anthropic. It raises a bigger question about who gets to decide how powerful AI systems can be used: governments, model companies, or both?

As AI moves deeper into defense, finance and government systems, those questions are likely to become more common.

AI-Assisted Hacking Is No Longer Just a Theory

Security researchers found that hackers used Cursor, a commercial AI coding tool, to help attack at least seven companies.

The important point is not that AI suddenly became a hacker. It is that attackers can use mainstream commercial AI tools to speed up parts of an intrusion.

That changes the economics of cybercrime. Tasks that previously required specialized people and significant time can increasingly be assisted by software.

Companies therefore need to think about AI as part of their security threat model, not simply as another productivity application.

Tencent Keeps Pushing Open-Source AI

Tencent released a preview of a new open-source AI model designed for software engineering, academic research and financial analysis.

The move reflects a broader trend in China's AI industry: competition is moving beyond model size toward practical capability, inference cost and developer ecosystems.

For developers, open models offer an important advantage: more control over deployment instead of relying entirely on a single hosted AI provider.