Christopher Kanan

Professor | AI Researcher

[Christopher Kanan @chriskanan]

I keep seeing “AI doomer” takes claiming today’s trajectory leads to the Terminator. It doesn’t. Much of the fear comes from abusing the term AGI. I explain why Path 1 ≠ Path 2, and why LLMs won’t “wake up” in this essay: https://t.co/yEgX119237

[CoLLAs 2026 @CoLLAs_Conf]

🎉 Announcing the winners of the #CoLLAs2025 Early-Career Spotlight Program!

These rising stars are pushing the boundaries of lifelong & continual learning, adaptation, and embodied AI.

👇 A thread introducing their work & talks.

#AI #MachineLearning #ContinualLearning #LifelongLearning

[Christopher Kanan @chriskanan]

UR released this Q&A with me, where I discuss my thoughts on the current state of AI, progress towards AGI, and why LLMs won't suffice for AGI: https://t.co/CPPQMhxQbF

#ai #agi

[Christopher Kanan @chriskanan]

Following my last essay about how AI will cause enormous reductions in college enrollments, I wrote a follow-up about what they can do to prepare: https://t.co/gRdwpYL27s

#ai #college #university

[The Academic Minute @AcademicMinute]

On @UofR Week: Human intelligence and artificial intelligence learn differently, but can that change?

Chris Kanan, associate professor of computer science at the Hajim School of Engineering and Computer Science, looks at the possibilities.

https://t.co/GFt9zf5P3H

[Christopher Kanan @chriskanan]

I was featured on the Academic Minute, which airs on over 70 radio stations around the US and Canada. In the segment, I introduce continual learning for the non-AI audience: https://t.co/3mynVySJXz

#ai #academicminute #continuallearning

[Christopher Kanan @chriskanan]

I've grown quite concerned about the future of academia and AI's impact on enrollments, as I explain in this essay: https://t.co/Uh3Ig58h5z

#ai #HigherEd

[Christopher Kanan @chriskanan]

@pmarca There is not enough forward-thinking regarding AI in higher education. At my university, I've argued that we are going to enter into an era where college is of decreasing value and the surviving universities will be those that have fully embraced AI.

[Christopher Kanan @chriskanan]

A proposed AI bill would, based on my read (and ChatGPT's), make it illegal in the USA to download AI code or weights created by Chinese companies, universities, etc. This is catastrophically shortsighted.

Given that roughly half of the academic AI papers published in our top-tier conferences are produced by Chinese universities, this would catastrophically impair AI research in the USA if researchers cannot download code or weights if they were developed by Chinese institutions.

https://t.co/m6HhXRgR5H

Proposed Bill's Text: https://t.co/QZLYs8WguK

[Christopher Kanan @chriskanan]

@elonmusk Given his role as a science communicator, it is depressing that Neil doesn't realize how critical it is to become a multi-planetary species. He will change his tone once it happens, just as he did regarding reusable rockets.

[CoLLAs 2026 @CoLLAs_Conf]

📢 Exciting News! The Fourth Conference on Lifelong Learning Agents (CoLLAs 2025) will be held at the University of Pennsylvania (@Penn) in Philadelphia, USA 🇺🇸

🗓️ Important Dates:

Abstract Deadline: Feb 21, 2025
Submission Deadline: Feb 26, 2025
Conference Dates: Aug 11 - Aug 14, 2025

We invite submissions that present new theories, methodologies, applications, or insights into algorithms and benchmarks designed for non-i.i.d. and non-stationary settings. Accepted papers will be published in the Proceedings of Machine Learning Research (PMLR). 📚

Full CFP: https://t.co/S8MwjLbOAr

#CoLLAs2025 #AI #MachineLearning #ContinualLearning #LifelongLearning #ResearchConference #CallForPapers #NonStationaryLearning

[Martin Mundt @mundt_martin]

Our Continual Causality Bridge is going into its 3rd edition at AAAI-25 @RealAAAI ! 🎉

Join us for a spectacular 2 days of continual learning & causality by submitting your original 4 page work (for proceedings or non-archival tracks) until Nov 25 AOE!

https://t.co/w9ghKcz5OL

[Christopher Kanan @chriskanan]

After years of effort, we have published Virchow, the first million whole slide image (WSI) foundation model for histopathology, in Nature Medicine! It excels at detecting rare cancers, and we are fervently building new models on top of its incredibly diagnostic and robust embeddings.

I'm super proud of the team, and it was a pleasure collaborating with the folks at Microsoft Research.

https://t.co/wYXWn1mWQg

[Christopher Kanan @chriskanan]

@giffmana @vikhyatk Did you write your own loader for it? I was recently asking why my lab's VQA datasets, which were made to be better than VQAv2, aren't being used for evaluating LLMs. I'm told it is because people are using LAVIS and its not in there

[Christopher Kanan @chriskanan]

@vikhyatk Really happy to see someone using one of these datasets my lab made for this application @giffmana . We spent a year making TallyQA and TDIUC because of all the weaknesses and biases in VQAv2, in addition to VQAv2 not enabling more refined analysis.

[Christopher Kanan @chriskanan]

My university made this lovely video about my work: https://t.co/To0oezqdZh

[Christopher Kanan @chriskanan]

We won a new $2M NSF award to develop methods that are more sample efficient for deep learning on temporal prediction and control problems! My lab is collaborating with the labs of Dhireesha Kudithipudi, Itamar Lerner, and Garret Rose for this exciting project.

https://t.co/zNiW12WUOe

[Christopher Kanan @chriskanan]

@ylecun @timoreilly This is an issue I've been raising in my classes and in talks. I'm worried about a handful of companies being in control of AI assistants resulting in cultural stagnation, especially as they are used by kids growing up.

[Christopher Kanan @chriskanan]

@karpathy This is what I tell my students and administration at the university. These are the very early days, and I think it will be far more transformative than they think in ways I probably can't even imagine. What I'm hoping for is "The computer" from Star Trek TNG

[Christopher Kanan @chriskanan]

@xwang_lk Certainly the lead author should review for the conference. My issue is that I often have collaborations where I helped design the method, but the paper is far from my own field for a conference I know nothing about, but they expect me to review.