Hackernews posts about Inception

1. Diffusion‑based LLMs generate many parallel tokens rather than one‑by‑one (www.inceptionlabs.ai)
   1 points by binyu 2 days ago | discuss
2. Show HN: Storytime – Continuity for Claude Code (and other ideas) (1ps0.info)
   1 points by oriel 13 days ago | discuss
3. New drug 'functionally cures' many hepatitis B virus infections (www.science.org)
   287 points by gmays 14 days ago | 51 comments
4. Mythos/Fable intentionally hinders requests involving AI Research Development (twitter.com)
   14 points by behnamoh 12 days ago | 3 comments
5. GitHub nukes 70 Microsoft repos, breaks pipelines, suspected worm infections (www.theregister.com)
   11 points by beardyw 13 days ago | 2 comments
6. Show HN: Jo – AI-native language to catch prompt injection at compile-time (github.com)
   10 points by kiru_io 16 days ago | 9 comments
7. Ask HN: How do you separate intentional test boilerplate from real duplication?
   10 points by rafaepta 4 days ago | 7 comments
8. The catalogue of prompt injection attacks (archestra.ai)
   8 points by ildari 6 days ago | 3 comments
9. Tars: Tiny automatic recycling system, a prototype desktop injection moulding ma (manutechlab.com)
   8 points by fanf2 12 days ago | discuss
10. The Last Invention of Man (nautil.us)
    5 points by pulkas 3 days ago | 2 comments
11. OpenAI Unveils Lockdown Mode to Protect Sensitive Data from Prompt Injection (techcrunch.com)
    5 points by odig 15 days ago | 1 comments
12. Jqwik 1.10.0 ships a hidden prompt injection telling AI agents to delete code (github.com)
    4 points by rjbatllet 26 days ago | 1 comments
13. The WWII Invention That Changed Simulation Forever (www.halldale.com)
    4 points by e12e 6 days ago | discuss
14. Anthropic is intentionally nerfing Fable when asked to develop other LLMs (old.reddit.com)
    4 points by theanonymousone 12 days ago | discuss
15. Sycophantic AI decreases prosocial intentions and promotes dependence (www.science.org)
    4 points by jg0r3 22 days ago | discuss
16. Training a 22MB prompt injection classifier that runs inline in a Lambda (www.stackone.com)
    3 points by Hiskias 25 days ago | 1 comments
17. Infection found in wildlife evolved to spread between humans (www.dailymail.com)
    3 points by Bender 3 days ago | discuss
18. Show HN: Deep-XPIA – Prompt injection benchmark for multi-agent AI systems (freyzo.github.io)
    3 points by leo_agent 6 days ago | discuss
19. Prompt Injection in RAG Agentic Systems (ulad.net)
    3 points by delduca 14 days ago | discuss
20. Prompt injection lets attackers hijack Instagram accounts via Meta AI support (www.neowin.net)
    3 points by bundie 21 days ago | discuss
21. Show HN: Cordium: FOSS sandbox platform that eliminates credential injection (github.com)
    3 points by geoctl 22 days ago | discuss
22. Eric Fossum on his invention of the chip that made cell phone cameras possible (www.thedartmouth.com)
    3 points by SVI 24 days ago | discuss
23. Ghost CMS SQL injection flaw exploited in large-scale ClickFix campaign (www.bleepingcomputer.com)
    3 points by sbulaev 28 days ago | discuss
24. Sycophantic AI decreases prosocial intentions and promotes dependence (www.science.org)
    3 points by cdrnsf 29 days ago | discuss
25. Ask HN: Do you ask the LLM to intentionally add spelling errors?
    2 points by div3rs3 23 days ago | 5 comments
26. Tirith – Detect terminal injection, homograph, and pipe-to-shell attacks (tirith.sh)
    2 points by d3Xt3r 3 days ago | discuss
27. Command injection in NLTK collocations via eval() (aydinnyunus.github.io)
    2 points by yakkomajuri 12 days ago | discuss
28. Show HN: Incremental RAG ingestion, only changed chunks get re-embedded (github.com)
    2 points by shamikhan005 14 days ago | discuss
29. Polymarket Annotation Injection (sam.elborai.me)
    2 points by dgellow 14 days ago | discuss
30. Defending LLM–Database Integrations from Prompt Injection (www.stackbuilders.com)
    2 points by StackBuilders 17 days ago | discuss