What Continual Learning Actually Is →
How models, agents, and agentic systems turn operating experience into persistent, verified improvement, with the challenges of generalization, adaptation, safety, and shared learning.
JP Vasseur · Research & writing
Building AI That Keeps Learning
Exploring continual learning across models, agents, and AI systems — how they learn from experience, adapt over time, and how we evaluate their progress.
Learn / Adapt / VerifyExplore the research →
Selected writing
Don't Imitate, Innovate: The Case for Useful AI
Judging AI by usefulness, contribution, and problem-solving impact.
Why AI Hallucination (Confabulation) is More Solvable Than Our Own
Why machine confabulation can be reduced more systematically than many human cognitive biases.
The Power-Hungry AI Myth
Why fears of an intrinsic drive for control misunderstand the architecture of current AI systems.
AI for networking
Beyond Protocols: Why AI is Networking's Overdue Paradigm Shift
How AI can move networking beyond individual protocols toward intelligent, distributed systems.
Time to Revisit the Internet Layering Principle: AI-Driven Cross-Layer Optimization
Using AI and multi-layer telemetry to optimize networks across traditional layer boundaries.
Research
Learning, intelligence, and the systems around them.
Current questions in continual learning sit alongside a longer body of work in AI, networking, and distributed systems.
Browse research and publications →Neuroscience
AI and Neuroscience: the (incredible) promise of tomorrow
A 2023 white paper exploring the connections between biological and artificial intelligence.
Read the white paper ↗
My Journey
The work behind the questions.
From internet architecture and AI for networking to today's research. Explore the books, talks, technical milestones, and personal interests along the way.
Explore my journey →Follow new writing
Get an email when I publish a new white paper.
Subscribe →RSS feed →