AI News, Jul 4, 2026: Anthropic launches drug discovery program for neglected diseases

Today's stories highlight AI's expanding footprint in healthcare, education, and corporate security, alongside growing regulatory scrutiny.

1. 01
   UK health service to use AI for patient triage on its appThe Guardian AI

   The National Health Service in England will deploy an AI tool on its official app to direct patients to appropriate medical services. The rollout is expected to reach approximately 200,000 patients over the next year as part of a major system overhaul.

   Why it matters

   Patients in England will interact with AI to find appropriate services, potentially reducing wait times but changing how they access care.
2. 02
   Australia warns doctors over privacy risks of AI medical scribesThe Guardian AI

   Australia's federal health department has raised concerns regarding the rapid adoption of AI scribes in general practice clinics. Regulators are monitoring the technology to address potential privacy pitfalls and ensure patient data remains secure.

   Why it matters

   Patients' sensitive medical discussions could be exposed if AI transcription tools do not meet strict privacy standards.
3. 03
   Midjourney demands Hollywood studios reveal their own AI usageTechCrunch

   In an ongoing legal dispute, image-generation startup Midjourney is seeking to compel three Hollywood studios to disclose details of their own AI usage.

   Why it matters

   The legal outcome could force major entertainment companies to disclose how extensively they rely on generative AI.
4. 04
   Alibaba bans employees from using Anthropic's Claude CodeTechCrunch

   Alibaba has reportedly classified Anthropic's Claude Code tool as high-risk software and banned its employees from using it. The decision highlights growing corporate anxiety over the security of AI-assisted programming tools.

   Why it matters

   Major tech firms are restricting AI coding assistants due to security concerns over proprietary code and potential data leaks.
5. 05
   Questions arise over OpenAI's promised multi-billion pound UK investmentThe Guardian AI

   OpenAI has reportedly failed to visit a key site for its proposed "Stargate UK" data center project, raising concerns that a highly publicized £20 billion investment may have been hypothetical.

   Why it matters

   The delay or absence of planned data centers could impact local tech infrastructure development and economic growth.
6. 06
   Large-scale study links AI homework help to lower exam scoresThe Decoder

   A study of over 26,000 students in China found that those using AI tools scored up to 24 percent lower on exams, despite completing homework faster. Researchers noted that the negative impact on learning took up to two years to fully surface.

   Why it matters

   While AI helps students finish homework faster, it can lead to a significant drop in actual exam performance over time.
7. 07
   Anthropic launches drug discovery program for neglected diseasesThe Decoder

   Anthropic has launched its own drug development program targeting neglected diseases that traditional pharmaceutical companies find unprofitable. The initiative aims to leverage AI to significantly reduce development timelines and improve clinical success rates.

   Why it matters

   It could accelerate treatments for unprofitable diseases, potentially cutting development times and doubling success rates.
8. 08
   Mistral releases open-source model that finds real-world software bugsThe Decoder

   Mistral AI released Leanstral 1.5, an open-source model designed for formal verification and mathematics. During testing, the model successfully identified five previously unknown bugs across dozens of open-source repositories.

   Why it matters

   Developers gain free access to a highly accurate model for verifying code and math, reducing software vulnerabilities.