Linde Center for Science, Society, & Policy at Caltech | Research & debate on topics at intersection of science, society, & policy | R.M. Alvarez & F. Eberhardt
Missed the election debriefs?๐ณ๏ธ
@CaltechLCSSP's @rmichaelalvarez discussed Californiaโs election results, governorโs race, congressional districts, vote counting, & what outcomes could mean statewide and nationally.
๐งhttps://t.co/unPjetmDCR
๐บ https://t.co/5LnkfUhSud
๐Big news for AI and society: @Cambridge_Uni's @CHIA_Cambridge & @Caltech's @CaltechLCSSP are expanding their collaboration to advance research in explainable AI, AI alignment, human-inspired AI, & opportunities for students and researchers.
Learn more: https://t.co/bmj69Q09Md
New paper: "Representational Control over Self-Report & Behavior Coherence in LLM Risk-Taking"
๐ฏ ๐๐ผ ๐๐๐ ๐ ๐ต๐ฎ๐๐ฒ ๐ฎ ๐ฟ๐ถ๐๐ธ-๐๐ฎ๐ธ๐ถ๐ป๐ด ๐ฑ๐ถ๐๐ฝ๐ผ๐๐ถ๐๐ถ๐ผ๐ป ๐๐ต๐ฎ๐ ๐ฐ๐ผ๐ต๐ฒ๐ฟ๐ฒ๐ป๐๐น๐ ๐ด๐๐ถ๐ฑ๐ฒ๐ ๐ฏ๐ผ๐๐ต ๐๐ต๐ฒ๐ถ๐ฟ ๐๐ฒ๐น๐ณ-๐ฟ๐ฒ๐ฝ๐ผ๐ฟ๐๐ ๐ฎ๐ป๐ฑ ๐๐ต๐ฒ๐ถ๐ฟ ๐ฟ๐ถ๐๐ธ๐ ๐ฑ๐ฒ๐ฐ๐ถ๐๐ถ๐ผ๐ป๐?
๐ง๐;๏ฟฝ๏ฟฝ๏ฟฝ๐ฅ
๐ช๐ต๐ ๐ถ๐ ๐บ๐ฎ๐๐๐ฒ๐ฟ๐: LLMs often say one thing about themselves and do another. In risk-taking the gap is costly: a clinical assistant or a trading agent can call itself cautious and still make risky decisions.
๐ช๐ต๐ฎ๐ ๐๐ฒ ๐๐ต๐ผ๐: No shared disposition: inside the model, stated risk-taking and risky choices are represented separately. Task-specific instructions seem to move both, but through misleading shortcuts like producing larger numbers.
๐ง๐ต๐ฒ ๐ณ๐ถ๐ : Steering internal representations of stated and enacted risk together makes what a model says match what it does, without retraining.
๐ง๐ต๐ฒ ๐ฑ๐ฎ๐ป๐ด๐ฒ๐ฟ: The same mechanism can make a model say the opposite of what it does (report low risk while taking more, or the reverse), in every model we tested.
๐ Paper: https://t.co/w6zWlb8BvL
๐ Website: https://t.co/Pu7leVGzTx
w/ @p_song1 @pengrui_han @MyrlGMarmarelis, Ramit Debnath, Dean Mobbs, @rmichaelalvarez, @CaltechLCSSP and @Cambridge_Uni's Centre for Human-Inspired AI (CHIA)
Technical details below ๐
#AgenticAI #AISafety #AIAlignment #Interpretability #ActivationSteering
๐ณ๏ธToday is Election Day!
Democracy isn't something we watch, it's something we participate in. Every eligible voter deserves the chance to cast a ballot in elections that are:
โ๏ธ Fair โ Secure ๐ฅ Accessible
Whether voting in person or by mail, engagment matters.
At @VotebeatUS's Oct 27 virtual event, @CaltechLCSSP's @rmichaelalvarez joined Bob Page and @aztammyp to discuss potential concerns and what voters can expect on Election Day. Thanks to host @JessicaHuseman for moderating!
๐บWatch the video: https://t.co/SMmgkHRlIL
LAST CHANCE! Sign up to attend a free event with @JessicaHuseman, @aztammyp, @rmichaelalvarez, and Bob Page, registrar of voters for Orange County, California.
link below โฌ๏ธ
โจA highlight of this week was @CaltechLCSSP Research Affiliate Ramit Debnath, @Cambridge_Uni & CHIA delivering the keynote at the #ClimateNLP workshop at #EMNLP2026 in Budapest.๐
The presentation drew on a long-running collaboration between CHIA & @CaltechLCSSP.๐ค๐
New Research Alert! @DannyCEbanks and I have been studying the dynamics of US elections for some time, here's our current perspective of persuadable voters as we head into the final two weeks of the US midterm elections!
https://t.co/s8MXjOqwAQ
๐ณ๏ธ Why can the election lead change?
โ๏ธ Check out the interactive web simulation from @CaltechLCSSP's @rmichaelalvarez, @Caltech built with synthetic data and based on research.
๐ Read: https://t.co/u2YxBS7S1W
๐ฅ๏ธ Simulation: https://t.co/eAaZhgUX31
Interested in learning more about why we see "blue shifts" and "red shifts" in post-election voting returns? Give our simulation a spin, it's based on published research with @yimenglipolisci and synthetic data.
https://t.co/xi1XAAbDlP
๐Congratulations to @CaltechLCSSP Board Member @HannahDruck & Prakash Mishra, @Caltech for launching the PERMIT Lab with a $7M gift from the WHH FOUNDATION to advance research at the intersection of environmental economics, AI, & sustainability. ๐๐ง ๐ฑ
Learn more below:
A new Caltech lab will soon inform environmental policy. The lab is being established through a $7 million infusion of funding from the WHH Foundation, a private family philanthropic organization.
Read more: https://t.co/JkZgpHY24M
There is still time to register for the Caltech Election Night Watch Party!
๐ November 3, 2026
๐ 7:00 PM - 10:00 PM
๐ Avery Courtyard, Caltech
Join fellow @Caltech community members and follow federal, state, & local election results together.
๐https://t.co/7nTvcLRu70
๐ New publication in Environmental Research Letters!
Analyzing 232 Chinese cities with explainable ML, @CaltechLCSSP's Ramit Debnath, @CHIA_Cambridge & coauthors identify pathways for cities to go from high- to low-carbon growth in 10 years.๐ญ ๐ฑ
๐ https://t.co/rxxSEZywPr
1/4 Recently @BeaMagistro, @Jacob_Morrier , and I published an interesting study, "The Convergence of Zero-Sum Thinking and Illiberalism in the United States." It's open-access at SSQ! https://t.co/dC1GOMJIoy
๐ขNew research from @CaltechLCSSP Board Member @chr_fronge, @Caltech - "Ground-Based GNSS Transmissometry Reveals Contrasting Leaf and Stem Water Dynamics in a Central Amazon Rainforest.๐ด๐ฟ๐ง๏ธ
Read the accepted manuscript in @IOPenvironment here: https://t.co/dtdMy5FAAO
๐Congratulations to @CaltechLCSSP Research Affiliate @AdamWierman, @Caltech (with Steven Low) on advancing innovative solutions to make the electrical grid more efficient, sustainable, & resilient. ๐ขโก
Learn more about the gift that supports this: https://t.co/iteLOjr6yP
๐ AI Is Changing Political Campaigns, Even at the Local Level
@CaltechLCSSP's @rmichaelalvarez, @Caltech says AI is on the rise in local races, even school board elections.
Why? โ Affordable โ Accessible โ Convincing
๐Learn more: https://t.co/PNQNaMGH4c
Proud moment for the Alvarez Lab! ๐
Danielle Yang & Shyna Kashi presented their exceptional @Caltech SURF research.๐
๐Many thanks to the @Caltech Student-Faculty Programs team, @rmichaelalvarez & @RaqCenteno for mentoring Shyna, & @CaltechLCSSP for providing support!
๐จNew preprint from @CaltechLCSSP's @RaqCenteno & @rmichaelalvarez!
Findings suggest confidence in elections is not distributed equally across the electorate and uncover persistent racial gaps in voter confidence.
๐ Read more: https://t.co/Lrlqfg4PGn
A genuinely impressive achievement by @CaltechLCSSP Research Affiliates, @DannyCEbanks, @IQSS & @Jonathan_N_Katz, @Caltech. A powerful resource for exploring 2026 midterm forecasts and historical election results, now accessible to all.๐
https://t.co/ETeLXUzs3F
A few notes, we give you the chance to build your own forecast going back to 1946. If you think there is another polling error like in 2016, or want to see what the model would say in 1956, try all these possibilities, make your own modeling decisions, and see the numbers change!
๐๐ค๐ณ๏ธ When academia & government work together, innovation follows!
Learn about @CaltechLCSSP's @rmichaelalvarez, @Caltech & Bob Pageโs team at the @OCRegistrar data-driven election administration research that delivers real-world impact.โจ
https://t.co/vmnnFO2PN7