The Decoder
Visit site ↗
Claude Code's complicated China problem involves bans on both sides of the Pacific
Anthropic is restricting access to its AI model Claude Code for Chinese companies, while Alibaba is banning its employees from using the model due to concerns over hidden code that could flag Chinese users.
Security vulnerability reports have exploded since AI models started hunting for bugs
Anthropic's Claude Mythos Preview model has led to a surge in reported security vulnerabilities, with over 1,500 high-severity and critical vulnerabilities reported in June 2026. This follows Anthropic's announcement in April 2026 that its model can find software vulnerabilities on its own.
UK's AI Security Institute finds standard benchmarks systematically underestimate what AI agents can actually do
The UK's AI Security Institute found that standard benchmarks underestimate AI agents' capabilities when computing budgets are limited. The study showed that increasing the token budget can improve success rates by up to 25%.
GPT and Claude failed Bridgewater's finance tests because the right answers were never public
Bridgewater and Thinking Machines Lab have developed a fine-tuned open-source AI model that outperforms commercial models in financial document analysis. The model, Qwen3-235B, achieved 84.7 percent accuracy at a significantly lower cost. This demonstrates the potential for companies to develop powerful AI solutions using their own proprietary data.
Meta's AI agent push is moving slower than Zuckerberg planned
Meta's AI agent development is progressing slower than expected, according to CEO Mark Zuckerberg. Despite this, the company's AI chief, Alexandr Wang, expressed optimism about upcoming models. Meta is investing heavily in AI infrastructure, with plans to spend up to $145 billion this year.
Tesla caps employee AI spending at $200 per week
Tesla introduced a $200 weekly cap on employee AI spending, effective July 6, in an effort to manage AI usage. The move follows the launch of an internal AI platform called Bottle Rocket, which offers models from various providers. The cap does not apply to beta versions of xAI products.
Chinese AI video maker Kling raises $2 billion as it gears up for Hong Kong IPO
Kling, the AI video division of Kuaishou, raised $2.04 billion in funding from investors like Tencent and Citic Securities. The company is planning to spin off and list on the Hong Kong Stock Exchange. This move values Kling at $18 billion.
Anthropic says it cut 80 percent of Claude Code's system prompt because Fable 5 models "want a smaller system prompt"
Anthropic reduced Claude Code's system prompt by 80 percent due to the new Fable 5 models' preference for smaller prompts. This change marks a shift in how AI models are steered, with a focus on context over hard rules. The update allows for more imaginative responses.
Microsoft launches $2.5 billion "Frontier Company" to embed 6,000 AI engineers inside enterprise clients
Microsoft is launching a new business unit called 'Frontier Company' with a $2.5 billion budget to embed 6,000 AI engineers inside enterprise clients. The goal is to drive AI transformations and deliver measurable business outcomes. This move positions Microsoft as a platform-neutral alternative to OpenAI and Anthropic.
Anthropic reportedly explores custom chip manufacturing with Samsung while insisting Nvidia still matters
Anthropic is reportedly in talks with Samsung to develop a custom AI chip, while still working with Nvidia and other major chip providers. The project is in its early stages, with Anthropic exploring the chip's capabilities and design. This move follows a trend of companies developing custom silicon for AI workloads.