-01.png)
Afterthoughts: AI Anxiety Session
- robinyap
- Apr 15
- 7 min read
The Problem: AI Panic Without a Playbook
Generative AI presents a unique dual challenge for L&D teams: like every other function, they face workflow automation risks and the potential obsolescence of their current product and service offerings but unlike other teams, they also carry the added responsibility of helping the broader workforce navigate AI-driven disruption.
Dr. Andrew Barrett opened the session by naming this tension directly. Drawing on the concept of the "Jagged Technological Frontier" which found that AI dramatically boosted consultant performance on tasks inside its capability zone (12.2% more tasks completed, 25.1% faster, 40% higher quality) while causing a 19-percentage-point drop in performance on tasks outside it Andrew framed AI disruption not as a uniform wave but as an uneven, unpredictable force. The session's core argument: the antidote to AI anxiety isn't prediction but having a repeatable, structured method for evaluating your offerings under uncertainty and a shared vocabulary that helps teams make defensible decisions and articulate their relevance to senior leadership.
The Framework: CATS in Action
The heart of the session was the ScaleLearning CATS AI Strategic Framework, a two-axis matrix that maps any L&D product or service along two dimensions: (1) Human Advantage (how much humans still outperform AI in delivering this offering) and (2) Future Viability (whether demand for the offering is likely to grow, hold, or erode).
Placement into one of four quadrants — Capitalize, Assess, Transform, or Shift — carries distinct strategic implications: upper-right (high human advantage, high viability) is lowest risk; lower-left (low human advantage, low viability) is highest.
Each quadrant poses a specific strategic question:
Can you maximize and expand the value of this offering?
Can you win the competition to automate and scale it?
Can you reinvent it to increase resilience?
Or should you pivot toward something more future-proof?
Participants worked through a hands-on application using a shared L&D portfolio plotting offerings like leadership development communities of practice and customer support e-learning onto the matrix then using the quadrant questions to draft the beginnings of a one-page action plan. The framework is designed to help L&D leaders move beyond AI panic toward a clear roadmap for relevance, upskilling, and reskilling and crucially, to produce an artifact that senior leaders will actually engage with.
The Bigger Picture: Why This Matters Now
Andrew closed with a provocation that resonates well beyond the session itself: teams that can't articulate the resilience of their own offerings won't be trusted to help others navigate change. The CATS framework isn't a one-time exercise but a planning muscle. Applied regularly, it positions L&D as a strategic function that leads with evidence rather than reacting with urgency. This aligns with a broader argument in the field: as Wharton professor Ethan Mollick argues in Co-Intelligence, the key isn't to fear AI or embrace it uncritically, but to actively experiment, find where AI's uneven capabilities help or hurt your specific work, and make deliberate partnership choices. Barrett's framework operationalizes exactly that instinct for L&D teams translating AI uncertainty into portfolio strategy, and portfolio strategy into leadership-facing action plans.
📚 Further Reading, Listening & Watching
From other experts to deepen and challenge the ideas from this session
📚 Books
Title | Author(s) | Who Mentioned | Context |
Co-Intelligence: Living and Working with AI | Ethan Mollick | Andrew Barrett | Andrew recommended it during the session |
The Leadership Pipeline | Charan, Drotter & Noel | Alex | Mentioned he was reading it after attending an information session the prior week |
The Coming Wave | Mustafa Suleyman | Andrew Barrett | Andrew referenced it after SHAD mentioned the Pi app — "I think it's called the upcoming wave or something like that, and I thought that was also a great read" |
AI and the Octopus Organization | Jonathan Brill & Steve Wunker | Rose Benedicks | Dropped into the chat during the human advantage discussion as "an interesting read" |
🎙️ Podcasts
Title | Host / Platform | Who Mentioned | Context |
The Amplifying Intelligence Podcast | — | Andrew Barrett | Shared in the pre-session chat: "a great one for L&D" |
Gilded Age | — | Alex | Pre-session chat: "my favorite" |
Huberman Lab — "Cultivating Awe" episode | Andrew Huberman | Urooj Mazhar | Pre-session chat, what she was listening to |
This is Love | Phoebe Judge | Tara Muir | Pre-session chat: "the podcast is excellent!" |
The Secret World of Roald Dahl | Spotify | Louise Platiel | Pre-session chat: "His life story is fascinating!" |
SPARK | CBC Radio 1 | Carroll | Recommended during the AI ethics/researchers discussion: "always on the edge of sci-fi in the real world" |
📄 Articles & Research Papers
Title / Description | Source | Who Mentioned | Context |
"AI Workforce Change: Slower Than Headlines Suggest" | Danielle Wallace | Shared as a link pre-session with the note "a lot of people are finding a lot of value in these insights" | |
"Navigating the Jagged Technological Frontier" (2023) | Dell'Acqua, Mollick et al. / Harvard Business School & BCG | Andrew Barrett | Core research underpinning the Jagged Frontier concept in the session |
METR research on AI task capability doubling | METR (Model Evaluation & Threat Research) | Andrew Barrett | Referenced to support the claim that AI capabilities have been doubling every 7 months |
🌐 Websites & Online Resources
Resource | URL | Who Mentioned | Context |
ScaleLearning CATS AI Strategic Framework | Andrew Barrett | This is available below. | |
Humanity's Last Exam / Safe AI benchmark | Ariel Harlap | Shared in response to Jennifer Cohen's question about AGI definitions | |
AIGS (AI Governance & Safety) | Robin Yap + Tara Muir + Khushroo Garda | Emerged during the ethics/researchers conversation as a relevant Canadian organization |
🏛️ Research Institutions & Organizations
Name | Who Mentioned | Context |
AMII — Alberta Machine Intelligence Institute (amii.ca) | Ariel Harlap | Canadian AI safety/responsibility research lab |
Mila (mila.quebec) | Ariel Harlap | Canadian AI research with ethics focus |
Vector Institute (Toronto) | Ariel Harlap | Same context — "each of them have different and overlapping takes on AI safety/responsibility" |
LawZero | Ariel Harlap | Mentioned as similar to AIGS, founded by Yoshua Bengio |
🤖 AI Tools & Apps Referenced
Tool | Who Mentioned | Context |
Pi (Personal Intelligence) by Inflection AI | Shad | Shared as a personal turning point |
Microsoft CoPilot | Jennifer Cohen | Noted it was the only AI tool widely adopted at her company |
Google Gemini | Khushroo Garda | Used to look up AIGS during the session |
👤 People & Thinkers Named
Person | Role | Who Mentioned | Context |
Ethan Mollick | Wharton professor, AI researcher | Andrew Barrett | Described as "a prolific writer and communicator on every single platform about AI"; Co-Intelligence recommended |
Tristan Harris | Tech ethicist, Center for Humane Technology | AD Blackwell | Raised in response to Tara's question about AI ethics experts |
Mustafa Suleyman | CEO Microsoft AI, co-founder Inflection AI | Andrew Barrett | Referenced in connection with the Pi app and The Coming Wave book |
Yoshua Bengio | AI researcher, Turing Award winner | Ariel Harlap | Named as the founder of LawZero |
Additional Materials you might find useful
📖 Books
Competing in the Age of AI — Marco Iansiti & Karim Lakhani (Harvard Business Press, 2020) By two of the researchers behind the Jagged Frontier study, this book examines how AI is restructuring the operating model of organizations — a useful strategic complement to Barrett's portfolio-level framework.
Navigating the Jagged Technological Frontier — Dell'Acqua, Mollick et al. (HBS Working Paper, 2023) The foundational academic paper introducing the "jagged technology frontier" — showing that AI assistance improves performance for some tasks while worsening it for others, even within the same knowledge workflow and at seemingly similar difficulty levels. Free to access on SSRN. A 30-minute read that radically reframes how to think about AI capability mapping — the empirical backbone of Barrett's Human Advantage axis.
🎙️ Podcasts
The Learning & Development Podcast — David James (360Learning) A fortnightly show hosted by David James, Chief Learning Officer at 360Learning and one of the top 10 global L&D influencers, with over 500,000 downloads — each episode debates topics affecting the profession alongside expert guests. Episodes like "How Generative AI is Transforming L&D Right Now" and "Rethinking L&D, Performance, and the Power of AI" (with Laura Overton and Donald Clark) directly extend the conversation Barrett started. One of the most credible practitioner-facing shows in the field.
Your Undivided Attention — Center for Humane Technology A December 2025 episode featured both Ethan Mollick and Molly Kinder of the Brookings Institution, who led research with the Yale Budget Lab examining AI's real-time impact on the labor market exactly the kind of macro-level evidence that contextualizes why Barrett's "Future Viability" axis matters. The show consistently brings rigorous, non-hype perspectives on AI's societal effects.
AI and the Future of Work — Dan Turchin (PeopleReign) Host Dan Turchin interviews thought leaders and technologists from industry and academia who share their experiences about artificial intelligence and what it means to be human in the era of AI-driven automation. Especially useful for CCCE members leading within organizations navigating workforce change, with strong episodes on reskilling, AI in HR, and the boundary between human judgment and algorithmic decision-making.
📺 YouTube / Video
Ethan Mollick — Co-Intelligence: An AI Masterclass (Stanford GSB, ~45 min) Mollick walks through the hype, fears, and potential of transformative AI, urging business leaders to experiment and discover rather than waiting on the sidelines. His practical "ten-hour rule" for building genuine AI fluency is a concrete counterpart to Barrett's call for strategic action over paralysis. Find it on the Stanford Graduate School of Business YouTube channel.
Powerpoint and Template provided at the session are here
Comments