Computer Architecture Today

Informing the broad computing community about current activities, advances and future directions in computer architecture.

From Theory to Practice: Introducing Architectural Prisms, an Experiment in AI-First Academic Dialogue

by Karu Sankaralingam on Nov 5, 2025 | Tags: , ,

From Theory to Practice: Introducing Architectural Prisms, an Experiment in AI-First Academic Dialogue

A little while ago, I published a post on this blog titled, “The Reviewer is Dead, Long Live the Review: Re-engineering Peer Review for the Age of AI.” In it, I argued that the traditional human-only peer-review system is buckling under the weight of rising submission volumes, reviewer fatigue, and misaligned incentives. I proposed that modern LLMs have reached a point of technical sophistication where an “AI-first” process could offer a more robust, objective, and tireless alternative.

Today, I’m excited to move that idea from a theoretical post into a practical experiment.

I invite you all to Architectural Prisms, a new, open platform for exploring and debating computer architecture research.

The concept is simple: just as a prism refracts a single beam of light into a full spectrum of colors, our platform uses AI to view cutting-edge research through multiple critical lenses. We are processing already-published papers from top conferences (starting with ISCA 2025 and ASPLOS 2025) and analyzing each one with three distinct AI personas:

  • The Guardian: Evaluates the technical rigor, methodology, and soundness of the work.
  • The Synthesizer: Places the research in its broader academic context, connecting it to related work.
  • The Innovator: Explores the potential for future impact and suggests new avenues for innovation.

Crucially, these AI-generated reviews are not verdicts; they are catalysts.

Our primary goal is to create a new public forum for critical discussion that goes beyond the traditional, high-stakes peer review or the confines of a classroom assignment. This is a place where the entire community—students, researchers, and practitioners—can critically examine what a paper contributes, where it falls short, how it can be built upon, and what new avenues it opens up. The code base is publicly available, as are the prompts.

This is an open experiment to help answer the big questions: How can AI best complement our human-intensive review process? Can it help us scale, remove bias, and improve the quality of academic dialogue?

The “reviewer” may not be dead yet, but the age of AI-assisted review is here. I hope you will join the experiment, read the reviews, challenge the AI, and lend your own human-led insights to the conversation.

You can join the discussion here: https://archprisms.talkyard.net/

Bio: Karu Sankaralingam is a Professor at UW-Madison and Principal Research Scientist at NVIDIA. He is an IEEE Fellow and holds the Mark D. Hill and David Wood Professorship at UW-Madison. He is a recipient of the Vilas Faculty Early Career Investigator Award in 2018, Wisconsin Innovation Award in 2016, IEEE TCCA Young Computer Architecture Award in 2012, the Emil H Steiger Distinguished Teaching award in 2014, the Letters and Science Philip R. Certain – Gary Sandefur Distinguished Faculty Award in 2013, and the NSF CAREER award in 2009.

Disclaimer: These posts are written by individual contributors to share their thoughts on the Computer Architecture Today blog for the benefit of the community. Any views or opinions represented in this blog are personal, belong solely to the blog author and do not represent those of ACM SIGARCH or its parent organization, ACM.