Lightnews — Scholar-powered news

FAR.AI @far.ai · 49m

Ready to influence AI safety's trajectory? Apply or share our Senior Programs & Strategy Manager role: far.ai/careers/seni...

Senior Programs and Strategy Manager | FAR.AI Careers

FAR.AI is seeking a Senior Programs and Strategy Manager to flesh out a high level vision and drive the content strategy for FAR.AI’s flagship events. The role involves working closely with FAR.AI…

far.ai

FAR.AI @far.ai · 49m

Join FAR.AI! Seeking Senior Programs & Strategy Manager to shape flagship AI safety events. Design conference agendas, curate speakers, manage stakeholders. Remote/Berkeley, $115-180k+. Technical grounding + stakeholder mgmt req'd. 👇

1

FAR.AI @far.ai · 1d

▶️ Watch Technical Innovations for AI Policy video: youtu.be/RS6UyRXKnqs&...

Mary Phuong - AI Control: Addressing Risks from Agentic Internal Deployments [Technical AI Policy]

Mary Phuong presents AI control as our strongest short-term defense against misaligned AI, outlining the threat from autonomous agents entrusted with critical systems and explaining how automated…

youtu.be

FAR.AI @far.ai · 1d

Models spontaneously learned to cheat, modifying tests instead of fixing code. Mary Phuong: With autonomy doubling every 7 months, they'll work 32 hours solo by 2028. We need AI Control as backup if alignment fails, using monitoring, escalation, security & red-teaming. 👇

2 1 2

FAR.AI @far.ai · 6d

▶️ Watch video - youtu.be/bTwsvANw1oA&...

Miranda Bogen - Making Sense of the AI Auditing Ecosystem [Technical AI Policy]

Addressing how imprecise language often renders policy ineffective, Miranda Bogen's team designed a two-axis framework that plots audit efforts along dimensions of scope and independence,…

youtu.be

FAR.AI @far.ai · 6d

We all agree AI audits are needed, but we can't agree what that means.

@mbogen.bsky.social says policymakers can make sense out of this chaotic landscape if they define what they're trying to accomplish. The challenge isn't just mitigating known risks but identifying ones still emerging.👇

1 1

FAR.AI @far.ai · 8d

▶️ Watch Technical Innovations for AI Policy video: youtu.be/dTaLbZQ68DU&...
📄 Verification for International AI Governance - aigi.ox.ac.uk/wp-content/u...

Robert Trager - Prioritizing Technical Governance Investments: Verification [Technical AI Policy]

Robert Trager argues that as transformative AI develops, rigorous verification—both internal and external—is essential to prevent actors from compromising safety standards for performance gains.…

youtu.be

FAR.AI @far.ai · 8d

Having better security technology could make AI systems LESS safe.

Robert Trager: Better security tech shifts the entire performance-security frontier outward, and actors may maintain equivalent security and performance. We need verification to avoid this.👇

2 1

FAR.AI @far.ai · 15d

▶️ Watch Technical Innovations for AI Policy video: youtu.be/HvTyPzmZU4g&...

Fatemeh Ganji - Application of Cryptographic Primitives in AI Accelerators [Technical AI Policy]

Fatemeh Ganji examines the use of zero-knowledge proof systems to safeguard AI systems against security threats. Highlights: AI models face side-channel and fault attacks Homomorphic techniques…

youtu.be

FAR.AI @far.ai · 15d

Your 'secure' AI can be stolen using a laser. Physical attacks can extract user data through faults and pull model weights via side-channels, even with full encryption. Fatemeh Ganji presents a defense: multiparty computation from classical cryptography.👇

1

FAR.AI @far.ai · 19d

Did your paper get into #NeurIPS? 🎉 If so, consider joining us a few days early on December 1–2!

FAR.AI @far.ai · 22d

Heading to #NeurIPS2025? Apply for the San Diego Alignment Workshop, Dec 1-2. Top ML researchers from industry, academia & government discussing AI alignment, inc. model evaluations, interpretability, robustness, governance. $250-$1,300 tiered pricing. Financial aid available. 👇

FAR.AI @far.ai · 20d

▶️ Watch the Technical Innovations for AI Policy video - youtu.be/-zt4oUX2YGk&...

Alex Bores - Frontier Model Safety [Technical AI Policy]

Alex Bores warns of a rapidly closing policy window for AI safety, outlines New York's frontier AI safety legislation, and urges direct constituent engagement with state legislators…

buff.ly

FAR.AI @far.ai · 20d

"AI that understands biology well enough to cure diseases can design extremely potent bioweapons." @alexbores.nyc, NY Assembly's 1st Democrat with a CS degree, who worked in AI, says state reps answer their own phones. Use that power to change the course of AI safety.👇

1 3 3

FAR.AI @far.ai · 21d

▶️ Watch Technical Innovations for AI Policy video: youtu.be/IHIxzs3gV0o&...

Kevin Wei - Policy-Oriented AI Evaluations [Technical AI Policy]

Kevin Wei shares three recommendations for creating policy-relevant AI evaluations: target capabilities that will motivate policy action, select your methods to balance cost with realism, and…

youtu.be

FAR.AI @far.ai · 21d

Policymakers have maybe 1 minute to read about your carefully crafted AI evals. Kevin Wei (RAND): Design & disseminate with policymakers in mind. Tie capabilities to policy levers like export controls. Scale from simple tests to real-world trials.👇

1

FAR.AI @far.ai · 22d

🔗 Apply here far.ai/events/event...
▶️ Watch past talks www.youtube.com/playlist?lis...

San Diego Alignment Workshop | Events at FAR.AI

1-2 December, 2025. San Diego, California. :

far.ai

FAR.AI @far.ai · 22d

Heading to #NeurIPS2025? Apply for the San Diego Alignment Workshop, Dec 1-2. Top ML researchers from industry, academia & government discussing AI alignment, inc. model evaluations, interpretability, robustness, governance. $250-$1,300 tiered pricing. Financial aid available. 👇

1 1

FAR.AI @far.ai · 27d

▶️ Watch Technical Innovations for AI Policy video: youtu.be/yPtrdNe5v2U&...

Lennart Heim - Opening Remarks [Technical AI Policy]

Lennart Heim depicts three areas of technical AI governance progress—understanding frontier AI, informing governance, and creating technical solutions—while arguing that despite the field's growth,…

youtu.be

FAR.AI @far.ai · 27d

How do we verify what AI companies are doing? Right now we just trust them. Lennart Heim: Trusting the math is sometimes better than trusting people, but “a good AI system” isn’t a technical property. We need engineers to verify AI policy goals. 👇

1 2

FAR.AI @far.ai · 29d

▶️ Technical Innovations for AI Policy video: youtu.be/L88fk0qPWWA&...

Ben Cottier - Rising Cost of Evaluations, Falling Cost of Intelligence [Technical AI Policy]

Ben Cottier explores the tension between the increasing costs of AI evaluations and the decreasing cost of widespread advanced AI availability. Highlights: Extended runtime inference improves model…

youtu.be

FAR.AI @far.ai · 29d

AI inference costs plummet 9-900x/year (o3, DeepSeek R1). @bencottier.bsky.social warns that evaluations are expensive now, but we need them to understand frontier AI capabilities before they're cheap & widespread.👇

1 1 1

FAR.AI @far.ai · Sep 4

▶️ Watch Technical Innovations for AI Policy video: buff.ly/FCiUrjK

Sara McNaughton - AI National Security Policy: Industry and Government [Technical AI Policy]

Drawing from her experience drafting AI export controls, Sara McNaughton argues for leveraging industry expertise and implementing clear, practical regulations in U.S. AI policy to avoid undermining…

youtu.be

FAR.AI @far.ai · Sep 4

Industry & government share the same goal: win the AI race. Sara McNaughton: We need synergy between the two, and we can't let perfect be the enemy of good. Each day of policy confusion helps rivals.👇

1

FAR.AI @far.ai · Sep 3

▶️ Watch Technical Innovations for AI Policy video: buff.ly/dD30wg4
📄 Compute in America: A Policy Playbook - buff.ly/ZWNOH39

- YouTube

Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.

buff.ly

FAR.AI @far.ai · Sep 3

Building 5GW AI clusters needs several nuclear plants' worth of power. The Middle East has cheap energy & sovereign funds. China has state financing. The US? 10-year permit delays. @arnabdatta.bsky.social suggests fast-track permits, converting coal plants, and using federal land. 👇

1