1. Bank of England Warns AI Could Accelerate Next Financial Crisis
International Business Times UK · Finance & Banking
The Bank of England’s July 2026 Financial Stability Report warns that rapid AI adoption across banking could amplify future financial crises by spreading market shocks faster through the system. The central bank flagged new cyber and operational risks from AI’s deep embedding in financial services, noting that frontier AI models can identify and exploit software vulnerabilities at greater scale. While AI brings benefits like improved fraud detection, the report cautions it could become a source of systemic risk.
https://www.ibtimes.co.uk/ai-financial-crisis-bank-england-warning-1809635
2. EU Commission Publishes AI Act Transparency Guidelines for Providers
European Commission · Law & Legal
On July 20, 2026, the European Commission published new guidelines on transparency obligations under the EU AI Act, targeted at providers and deployers of certain AI systems. The move is part of the Commission’s broader effort to clarify compliance requirements as the AI Act moves toward full applicability on August 2, 2026. The guidelines aim to give businesses and national authorities clearer direction on how to meet the law’s disclosure and transparency rules.
https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai
3. AI Drug Discovery Faces Pivotal Test as 15-20 Programs Enter Phase III
AIM Media House · Healthcare & Pharmaceutical
With roughly $60 billion invested in AI drug discovery since 2019 and over 175 AI-originated drug programs in human trials, 2026 is shaping up as the make-or-break year for the sector — none has yet received FDA approval. Industry estimates suggest 15 to 20 AI programs could enter pivotal Phase III trials this year, including candidates from Isomorphic Labs, which raised $2.1 billion in May. Analysts place roughly a 60% chance on the first AI-designed drug receiving regulatory approval by 2027, contingent on clinical data generated this year.
https://aimmediahouse.com/ai-lifesciences/2026-is-the-year-ai-drug-discovery-meets-clinical-reality
4. DARPA and US Air Force Fly AI-Controlled F-16 in Landmark Autonomous Combat Test
Stars and Stripes · Defense & Security
The US Air Force and DARPA have successfully flown a legacy F-16 Fighting Falcon under AI control at Eglin Air Force Base, Florida, as part of the VENOM program — a major milestone in autonomous combat aviation. A safety pilot remained in the cockpit and could toggle between human and AI control with a single switch, while the AI managed the aircraft for the majority of the flight. The breakthrough paves the way for human pilots to command teams of autonomous, uncrewed aircraft in future combat operations.
5. US Weighs Independent Body to Evaluate Frontier AI Models Pre-Release
Tech Startups · Technology & AI
The US government is considering creating an independent organisation, modelled on financial regulators like FINRA, to evaluate leading AI models before or after their public release. The proposed body would establish testing procedures, review safety claims, and assess model capabilities — providing technical expertise that federal agencies may struggle to maintain internally. Key details such as the body’s legal authority, funding, and independence remain unsettled.
6. NHTSA Issues Ultimatum to Robotaxi Firms Over First Responder Failures
TechCrunch · Transportation & Logistics
The US National Highway Traffic Safety Administration has issued a formal call to action demanding that autonomous vehicle developers fix a recurring pattern of their vehicles interfering with emergency services, setting a deadline of end-of-July for all AV companies to present solutions. The directive follows high-profile incidents including Waymo robotaxis being stranded during a July 4 traffic jam in San Francisco, prompting a local supervisor to launch a formal inquiry. The regulator was clear that failures to detect and respond to emergency scenes are not edge cases but fundamental safety deficiencies.
https://techcrunch.com/2026/07/12/techcrunch-mobility-a-robotaxi-ultimatum/
7. Virginia Law Triggers AI Policy Race Across US School Districts
VCU News · Education
A new Virginia state law that took effect on 1 July 2026 has directed the state’s Department of Education to create formal guidelines on AI use in public schools, with local school boards eventually required to adopt their own policies. New research from Virginia Commonwealth University found that while teacher and student AI usage has grown sharply, guidance and training are badly lagging behind adoption trends. The development mirrors a national wave of AI governance activity, with 134 education AI bills introduced across 31 US states in 2026 alone.