1. Only 21% of Banks Actually Measure AI’s Revenue Impact
FinTech Global · Finance & Banking
A new nCino benchmark report published on 28 August 2026 reveals that while 91% of banks now have an AI strategy in place, only 21% are measuring what that AI actually delivers to the bottom line. The research also found that 81% of banking executives say their organisations prioritise AI adoption over return on investment, highlighting a widening gap between deployment ambition and financial accountability.
https://fintech.global/2026/08/28/only-21-of-banks-measure-ais-revenue-impact/
2. Federal Judge Rules Pentagon’s Blacklisting of Anthropic Illegal
Forbes · Law & Legal
On August 28, 2026, U.S. District Judge Rita Lin ruled that the Pentagon’s designation of AI company Anthropic as a national security supply-chain risk was unconstitutional, finding it violated the First and Fifth Amendments as unlawful retaliation for the company’s public criticism of military AI use. The dispute began when Anthropic refused to remove safeguards preventing its Claude model from being used for mass surveillance or autonomous weapons. The 59-page ruling permanently blocks federal agencies from enforcing the blacklisting and marks a major legal victory for Anthropic ahead of its anticipated IPO.
3. First Fully AI-Designed Drug Rentosertib Enters Phase III Lung Disease Trial
GxP News · Healthcare & Pharmaceutical
Insilico Medicine has launched a Phase III clinical trial for rentosertib, an AI-discovered and AI-designed drug targeting idiopathic pulmonary fibrosis — a fatal, progressive lung disease with few effective treatments. It is the first time in medical history that a molecule fully designed by generative AI, targeting a target also discovered by AI, has reached late-stage pivotal testing. The randomised, placebo-controlled trial aims to enrol 320 patients across 47 centres in China over 52 weeks.
4. Time Magazine Tests Ads Targeting AI Crawlers, Not Human Readers
Saurce (UK) · Media & Marketing
Time has begun testing a new form of advertising designed specifically for AI crawlers rather than human readers, with brands like Ally Bank among the first advertisers — aiming to shape how AI systems understand and describe those brands. The experimental format raises a fundamental question for marketers: when an AI assistant helps someone research and choose a product, are brands marketing to the customer, the AI, or both? Perplexity has already pushed back, calling bot-only content a form of cloaking, signalling that the approach is controversial across the industry.
https://www.saurce.co.uk/insights/ai-in-marketing-takeaways-for-leaders-august-2026
5. UK Actors Demand Legal Protection Against AI Voice Cloning
Tech Startups · Technology & AI
A group of around 80 UK actors and public figures, including Nicola Coughlan and Hugh Bonneville, have signed an open letter urging the British government to legally recognise a person’s voice as a protected part of their identity. The ‘Save Our Voices Now’ campaign warns that AI can now clone someone’s voice from just a few seconds of audio, enabling fraud, impersonation, and deepfake content without consent. Campaigners cite data showing 28% of UK adults have already encountered voice-cloning scams.
6. Agriculture IoT Market Hits $8.86B as AI-Driven Precision Farming Accelerates
GlobeNewswire / MarketsandMarkets · Agriculture & Food
A new MarketsandMarkets report published August 25 values the global agriculture IoT market at $8.86 billion in 2025, projecting growth to $12.61 billion by 2030. The expansion is fuelled by rising demand for AI-powered drones, sensors, and automated machinery that improve crop yields and reduce waste. Climate variability and surging global food demand are driving adoption of precision farming as a foundational technology.
7. Palo Alto Networks: AI-Authored Malware Rising but Not Evading Detection
Palo Alto Networks Unit 42 · AI Cybersecurity
Palo Alto Networks’ Unit 42 research team has published new findings on AI-enabled malware, analysing 405 unique samples collected from threat intelligence sources ahead of Black Hat 2026. The report concludes that while AI is changing how malicious code is authored — making it faster to produce — it does not currently help malware evade existing behavioural detection systems. Enterprises are nonetheless urged to review SOC workflows as AI accelerates adversarial reconnaissance and payload delivery at scale.
https://unit42.paloaltonetworks.com/ai-enabled-malware-analysis/