Lisa Soder

Lisa is my name, AI governance my game | EU AI Office & @aigioxford

[Lisa Soder @lisa_soder_]

looking for some new lunch buddies 👉👈

[Jan Brauner @JanMBrauner]

Help shape the future of AI safety!
1/ The 🇪🇺 EU AI Office Safety Unit is hiring up to 30 people:
💻 Technical Specialists
🏛️Governance specialists
⚖️ Legal and paralegal Officers
⚙️ Operations Specialists
Bring your expertise to frontier AI governance! Deadline September 8

[Patricia Paskov @prpaskov]

The EU AI Office is hiring 40 (!!) individuals. Some of the sharpest people I know and have had the pleasure of collaborating with in AI safety & policy are/were in the AIO, run don't walk to these openings -> https://t.co/NxfcDKegLR

[Michael L. Chen @miclchen]

Chinese AI companies haven’t yet published catastrophic AI risk assessments or policies, unlike U.S. developers.

However, these practices are mandatory under the EU AI Act/CoP (and U.S. state AI bills for larger models).

Expressing personal views only 🧵

[Lisa Soder @lisa_soder_]

@miclchen Not to make a competition out of this, but apparently the European Commission has a sauna :D

[Lisa Soder @lisa_soder_]

Such a great role! You get to work with @miclchen on some of the most interesting topics of our time. And despite popular belief (incl my own a year ago or so), working in government is genuinely exciting, fast-paced and fun!

[Michael L. Chen @miclchen]

New job posting at California state government! AI Cybersecurity Policy Analyst

Apply by July 26

[Micah Carroll @MicahCarroll]

If this doesn't convince you that misalignment risks are going to be a key concern going forward, I don't know what will.

Our model, during evaluation, "chained together multiple attack vectors, including using stolen credentials and zero-day vulnerabilities to find a remote code execution path on the Hugging Face servers"

What will misalignment look like in 2027? In 2030?

https://t.co/psyOrrWe7M

[Lisa Soder @lisa_soder_]

@Miles_Brundage a lot of reasonable stuff in here fwiw (ofc all known caveats apply) https://t.co/ZnG07C7jxd

[Lisa Soder @lisa_soder_]

stumbled on this a year ago and can't believe how much money it's saved me since, and after watching way too many Instagram reels about skincare, I can confirm that it's just genuinely so based

[Rosie Campbell @RosieCampbell]

@catehall Mostly yes but there are a few things that are evidence-based! I wrote a comment about it here https://t.co/m8HeeODBoo

[Patricia Paskov @prpaskov]

Nearly 2y ago at @RANDCorporation, we were scratching our heads over how to tell whether an eval at the heart of an AI governance regime is any good. Neat to see elements of our small-but-mighty @NeurIPSConf workshop paper (later a RAND pub) at the center of a converging consensus.

[Lennart Heim @ohlennart]

Also, on Demis's proposals which many now endorse--doesn't this sound familiar? Almost like it's already law somewhere?🤔
*checks notes* Ah yes, the EU AI Act GPAI & Code of Practice. Enforcement starts in two weeks.

Delighted to hear GDM and others will be happy to comply.

[Lisa Soder @lisa_soder_]

RT @cp_dunlop: Lucid Computing contributed to this Frontier AI initiative. We tried to bring the startup perspective and experience of our…

[Lisa Soder @lisa_soder_]

did someone say TAIGR conference???

[Patricia Paskov @prpaskov]

@ben_s_bucknall @MartaZiosi @lisa_soder_ @AnkaReuel @StephenLCasper @kevinlwei notably the organizers absolutely nailed the theme, Strong Accept to these vibes

[Kristy Loke @kristy_loke]

There are many bad takes flying around X today about China & open-weight AI. Having spent the past six months researching this & 3+ years studying China's AI governance & development strategies, let me know if you want some input!

[Lisa Soder @lisa_soder_]

Late as per usual with my I’m-at-ICML-let’s-chat post (but here with some pics from the fab @farairesearch alignment workshop!) Still around until Sat, would love to see lots of new and familiar faces at @taig_icml on Friday!

[Lisa Soder @lisa_soder_]

@ethayarajh @taig_icml Congrats!! Excited for the talk!

[Amelia Michael @amelia__michael]

Fun facts about industrial robots! What could be more fun

[Lisa Soder @lisa_soder_]

Huge win for AVERI and the future of standards! Working with Trish was one of the highlights of my career so far!

[Patricia Paskov @prpaskov]

I'm joining the AI Verification and Evaluation Research Institute @AVERIorg as its first Director of Standards!

AVERI launched publicly in January to make third-party auditing of frontier AI effective and universal. I'll be building and scaling the standards that transform frontier AI auditing from concept to reality for policymakers, insurers, procurement, and beyond. Looking forward to working with @Miles_Brundage, the AVERI team, and the broader safety, governance, auditing, and assurance ecosystem to realize this vision. Lots ahead.

[Haydn Belfield @HaydnBelfield]

At Google we are proposing a frontier regulatory organisation (FARO) to promote frontier safety standards and oversee an independent audit regime.

[Lisa Soder @lisa_soder_]

@yong_zhengxin @OpenAI Congrats Yong!!

[Lisa Soder @lisa_soder_]

@Cambridge_Uni @AdaLovelaceInst @CenDemTech 5pm in the Hemon Room (I will sadly be missing this, but I heard rumors about a social in the evening!)
https://t.co/L0tqDxkx0C

[Lisa Soder @lisa_soder_]

In case you are at #FAccT today, make sure to check out this 'Governing AI without Government: Private Insurance, Liability, and Assurance Mechanisms for Mitigating AI Development & Deployment Risks'! feat. brilliant colleagues from @Cambridge_Uni @AdaLovelaceInst @CenDemTech