The AI Pulse

Issue #49 — 6th October2025

Editor: Professor Alan Brown

This week we recommend you download a great White Paper “The AI Teammate Framework: A Four-Step Framework for Product Teams” from Scrum.org. The paper tackles the challenge many face with wasting money and achieving poor results from random AI experimentation. Click the banner below to download a copy.

Highlights in this edition include:

UK government recovers a record £480m using AI fraud detection tool - Fraud Risk Assessment Accelerator to be licensed internationally.

AI-generated “workslop” costs companies nearly 2 hours per incident - Harvard study reveals 41% of workers encounter low-quality AI output requiring rework.

73% of organizations globally integrate AI into cybersecurity strategies - The financial services sector leads at 82% adoption rate.

Half of the US workforce uses AI tools without knowing if it’s allowed - 44% knowingly use AI improperly at work.

UK introduces a comprehensive AI testing framework for the public sector - First government-backed standardized approach to AI evaluation.

AI for Good

US investigators using AI to detect child abuse images made by AI - MIT Technology Review reports law enforcement deploying AI systems to identify AI-generated child exploitation material in a critical child safety application.

New research paves the way for safe and responsible AI in defence - Alan Turing Institute announces a research framework to ensure AI deployment in defense contexts maintains safety and ethical standards.

Bias and Ethics

Tim Berners-Lee says today’s internet exploits users for data - Harvard Crimson reports that the web inventor warns of surveillance capitalism and calls for digital rights protections.

New Playbook Offers Guidelines on Advancing Transformative Responsible AI - TechRepublic reports World Economic Forum and Accenture have released a framework for implementing AI with ethical safeguards and accountability.

Cyber Security

The future of payments research 2025: AI everywhere. Trust nowhere? - The Payments Association highlights growing tension between AI adoption in financial services and consumer trust concerns.

CISOs advised to rethink vulnerability management as exploits sharply rise - CSO Online reports that security leaders must adopt new approaches to vulnerability management as exploit activity accelerates beyond traditional mitigation strategies.

Data & Decision Making

The UK’s National Data Strategy must evolve to unlock AI in the public sector - Computer Weekly argues that public sector AI adoption requires fundamental rethinking of data governance and sharing frameworks.

The Modern Data Stack for AI: Engineering Foundations That Scale - The Data Scientist outlines infrastructure requirements for organizations building AI capabilities from startup to enterprise scale.

DPI for AI (and for properly realising the promise of digital) - Apolitical explores digital public infrastructure frameworks necessary to realize AI’s potential for public services.

AGI Benchmarks: Tracking Progress Toward AGI Isn’t Easy - IEEE Spectrum explores challenges in measuring progress toward artificial general intelligence amid competing evaluation frameworks.

Mapping relevant data collection mechanisms for AI training - OECD outlines frameworks for organizations to collect and manage data appropriately for AI model training while addressing quality and compliance considerations.

Innovation & Collaboration

CEO strategies for leading in the age of agentic AI - McKinsey outlines leadership approaches as autonomous AI agents reshape organizational decision-making and workflows.

The six key elements of agentic AI deployment - McKinsey shares practical lessons from organizations deploying autonomous AI systems at scale.

Perplexity’s Comet browser available to everyone, everywhere, for free - The Verge reports AI-powered browsers are becoming widely accessible, signaling a shift in search and browsing experience.

MP becomes first to create himself as an AI bot - BBC reports that a UK parliamentarian is deploying an AI constituent service tool, testing boundaries of automated representation.

Productivity & Efficiency

The AI Slop Era Arrives - Stratechery warns that the proliferation of low-quality AI-generated content threatens to undermine productivity gains and information quality.

Real AI Agents and Real Work - Ethan Mollick and One Useful Thing examines the tension between human-centered work and AI-generated output, questioning true productivity impacts.

Unlocking AI’s full potential requires operational excellence - MIT Technology Review argues that successful AI deployment depends on organizational readiness beyond technology implementation.

AI Leaders Outpace Laggards with Double the Revenue Growth - BCG reports that the top 5% of “future-built” companies achieve 1.7x revenue growth through comprehensive AI transformation.

Regulation and Compliance

Rethinking Economics Assessments for a GenAI World - The Royal Economic Society has published guidance on adapting academic evaluation methods in the GenAI era.

Harmonizing AI Guidance: Distilling Voluntary Standards and Best Practices into a Unified Framework - Georgetown’s Center for Security and Emerging Technology synthesizes fragmented voluntary AI standards into a cohesive framework for organizations navigating multiple guidance documents.

Generative Artificial Intelligence in Qualitative Data Analysis - Sage Journals examines methodological implications of using AI for research analysis versus traditional approaches.

The impact of digital technologies on students’ learning - OECD research examines how digital tools affect educational outcomes, providing evidence-based insights for policymakers and educators.

Sustainability

Responding to the climate impact of generative AI - MIT News explores strategies to address the energy consumption and environmental costs of AI infrastructure expansion.

What is alert fatigue and how can AI reduce it? - TechRadar explores how AI can help reduce information overload in network monitoring and management systems.

How AI Data Centers Are Sending Your Power Bill Soaring - Bloomberg analysis reveals that the AI infrastructure boom is driving up electricity costs for consumers across multiple regions.

AI and our energy future - MIT Technology Review presents a comprehensive package examining the intersection of AI development and energy demands, exploring infrastructure challenges and potential solutions.

User Experience

Perplexity’s Comet browser is available to everyone, everywhere, for free - The Verge reports that the AI-native browser democratizes access to AI-powered search and browsing capabilities.

Generative AI and Labour Market Outcomes: Evidence from the United Kingdom - SSRN research analyzes early indicators of GenAI’s impact on UK employment patterns and wage dynamics.

Workforce & Skills

Accenture plans on ‘exiting’ staff who can’t be reskilled on AI - CNBC reports that the consulting giant is cutting 11,000 roles as part of $865M restructuring to prioritise AI capabilities.

AI could impact a million jobs in London, research suggests - BBC analysis indicates significant workforce transformation ahead for UK capital as AI adoption accelerates.

US jobs market yet to be seriously disrupted by AI, finds Yale study - The Guardian reports that Yale University Budget Lab research shows labour market stability 33 months post-ChatGPT, though disruption may still emerge.

The AI talent war is becoming fiercer - The Economist examines the intensifying competition for AI expertise as organizations rush to build capabilities.

Have a comment or a story you’d like to share with us?

Just get in touch at: [email protected]

Do you know anyone who might be interested in AI Pulse

Refer a Friend

The AI Pulse is a free newsletter for Digital Leaders members. Pledge your support.