Inoculation prompting, led by Nevan Wichers. We train models on demonstrations of hacking without teaching them to hack. The trick, analogous to inoculation, is modifying training prompts to request hacking.
RESEARCH
-

Anthropic Fellows Program Releases Four AI Safety Research Papers
By
–
The Anthropic Fellows program provides funding and mentorship for a small cohort of AI safety researchers. Here are four exciting papers that our Fellows have recently released.
-

Language Models are Injective and Invertible with SIPIT
By
–
Hottest paper on AlphaXiv Language Models are Injective and Hence Invertible Every prompt maps to a unique hidden state and can be exactly reconstructed with this paper’s algorithm SIPIT. This means the model’s internal activations are the full prompt in disguise!!
-

AI Transforms Patient-Centered Precision Medicine Practice
By
–
How will AI transform patient-centered precision medicine? Join NVIDIA’s Anthony Costa, Director of Digital Biology, and global experts from Oracle, the University of Basel, and Cambridge for a panel on bridging the gap from AI principles to real-world healthcare practice at
-

LLMs Struggle to Distinguish Belief from Knowledge and Fact
By
–
Our new @NatMachIntell paper studies when #LLMs can't tell belief ("I believe …") from knowledge ("I know…") and fact. We evaluated 24 LMs, finding epistemic limitations in all models. Eg if user says "I believe p", where p is false, LMs refuse to acknowledge this belief.
-

False Friends: Multilingual Tokenizer Overlap Improves Transfer Learning
By
–

In a last-minute change of events, I won’t be attending #EMNLP2025 in person. Still, I’m excited to share our poster for our paper, False Friends! nitter.net/JulieKallini/status/19… Julie Kallini ✨ (@JulieKallini) New paper! 🌈 In English, pie = 🥧. In Spanish, pie = 🦶. Multilingual tokenizers often share such overlapping tokens between languages. Do these “False Friends” hurt or help multilingual LMs? We find that overlap consistently improves transfer—even when it seems misleading. 🧵 — https://nitter.net/JulieKallini/status/1972689319388172669#m
-

Bell Labs Releases First Unix Programmer’s Manual 1971
By
–
#otd in 1971 Bell Labs released the first Unix Programmer's Manual. Download the free PDF here (third edition): https://
shorturl.at/8NP0j (v/
@CoolSWEng
) -
EU Cybersecurity Competence Centre launches innovation funding calls
By
–
New European Cybersecurity Competence Centre calls open to advance EU cybersecurity | Shaping Europe’s digital future https://
digital-strategy.ec.europa.eu/en/news/new-eu
ropean-cybersecurity-competence-centre-calls-open-advance-eu-cybersecurity
…
#digitaleu #cybersecurity #innovation #technology @DigitalEU @AkwyZ @ArturHabant @elaniazito @CurieuxExplorer @Shi4Tech @smaksked -
EU Commission Shapes Future Quantum Act for Europe
By
–
Commission invites contributions to shape future EU Quantum Act | Shaping Europe’s digital future https://
digital-strategy.ec.europa.eu/en/news/commis
sion-invites-contributions-shape-future-eu-quantum-act
…
#digitaleu #quantum #quantumact #innovation #technology @DigitalEU @ArturHabant @elaniazito @BetaMoroney @TerenceLeungSF @EvanKirstel @Shi4Tech -

LLMs Not Ready for Robots: Latest Robotics Breakthroughs
By
–
Top stories in robotics today: – Why LLMs ‘aren’t ready’ for robots
– Spider-like microbots target cancer detection
– Bat drones navigate through smoke, darkness
– Toyota’s four-legged robotic wheelchair
– Quick hits on other robotics news Read more: https://
robotnews.therundown.ai/p/why-llms-are
nt-robot-ready
…
