Congrats to LIDS Faculty PIs Tamara Broderick and Caroline Uhler, elected to the IMS Fellowship for their outstanding research and professional contributions. They will be honored at the Bernoulli Society-IMS 11th World Congress in August bit.ly/4aDmKNC

Congrats to LIDS PI Ashia Wilson and her students Shomik Jain and Vinith M Suriyakumar on the best paper ACM FAccT!

What Can AI Really Do to Fight Climate Change? LIDS faculty PI Priya L. Donti talks about her work to identify climate applications for AI and why it’s in everyone’s interest to actively shape AI’s development on Bloomberg Zero Listen: spoti.fi/4eej2wN bloom.bg/3Xdxhw3

Algorithms and Behavioral Science, a new course led by LIDS PI Ashesh Rambachan and 𝐒𝐞𝐧𝐝𝐡𝐢𝐥 𝐌𝐮𝐥𝐥𝐚𝐢𝐧𝐚𝐭𝐡𝐚𝐧, helps students investigate the deployment of machine-learning tools and their potential to understand people, reduce bias, and improve society. Learn more: bit.ly/4eoBvXz

Launched in 2001 by LIDS PI & former MIT School of Engineering Dean Thomas Magnanti, MIT’s Undergraduate Practice Opportunities Program is a yearlong career-development course for 2nd-year students that uses experiential learning to bolster professional development. bit.ly/3xwJYaH

Why are AI models that analyze medical images biased? A new study from LIDS PI Marzyeh and MIT researchers explores why AI models that are the most accurate at predicting race and gender from X-ray images also show the biggest “fairness gaps.” bit.ly/3VMBEvs

LIDS PI Luca Carlone is among eleven MIT School of Engineering faculty to receive tenure in 2024. Carlone’s research is focused on the cutting edge of robotics and autonomous systems. Congrats Luca Carlone! Learn more: bit.ly/45O3uw9

When should you trust an AI model? IF-COMP, a new technique from LIDS researchers Nathan Ng and Marzyeh Ghassemi could help people determine whether to trust an AI model’s predictions. Learn more: bit.ly/4cSAC8R

How to assess a general-purpose AI model’s reliability before it’s deployed. A new technique from MIT LIDS researchers Navid Azizan and Young-Jin Park enables users to compare several large models and choose the one that works best for their task. bit.ly/3zNWiEf

A new AI model developed by an interdisciplinary team of researchers from MIT EECS, MIT LIDS, MIT IDSS and ETH Zurich, could help clinicians assess breast cancer stage and ultimately help in reducing overtreatment. news.mit.edu/2024/ai-model-…

LLMs don’t behave like people, even though we may expect them to. A new study from LIDS PIs Ashesh Rambachan and 𝐒𝐞𝐧𝐝𝐡𝐢𝐥 𝐌𝐮𝐥𝐥𝐚𝐢𝐧𝐚𝐭𝐡𝐚𝐧 shows someone’s beliefs about an LLM play a significant role in the model’s performance and are important for how it is deployed. bit.ly/3y2zmRm

Fairness may require some sort of randomization. A study from PI Ashia Wilson and grad student Shomik Jain shows that when allocating scarce resources with AI, randomization can improve fairness. More at MIT News: bit.ly/46v3S2K

MIT LIDS PI Dr. Caroline Uhler, grad student Xinyi Zhang, and ETH Zürich collaborators developed an AI model that can identify certain breast tumor stages likely to progress to invasive cancer. The model could potentially help to reduce overtreatment. bit.ly/3SK5NuS

A recent study led by researchers in MIT LIDS, shows that when a large language model is misaligned with a person’s beliefs, even an extremely capable model may fail unexpectedly when deployed in a real-world situation.news.mit.edu/2024/large-lan…

LIDS PI Kalyan Veeramachaneni and his Data-to-AI Lab developed a new LLM to detect anomalies in data recorded over time, without the need for training. The new method could help alert technicians to potential problems in equipment like wind turbines or satellites. bit.ly/4dMyDD2

New LIDS PI Gioele Zardini discusses his work and teaching in a spotlight for MIT CEE. Learn more about his research addressing the design challenges of “complex systems” and the new class he's teaching this fall. bit.ly/4dQv2Eo