3 research idea notes tagged “security”. Read them as Markdown, JSON or an Atom feed.
Prompt injection, jailbreak attacks and LLM security appear on 37 recent papers, all in AI journals. Banks and brokers are deploying agents that read emails, documents and web pages and then act. A hidden instruction in a document can make such an agent move money or leak data. How exposed are deployed financial agents, and which defences hold?
Papers in the last 30 days — prompt injection: 1
Financial fraud appears on 48 recent papers, but most study fraud by firms or fraud detection in transactions; only 6 carry a social-media, crypto, elderly or victim keyword, and "fraud victimization" on 5. Investment scams run through social platforms now cost households billions. Who falls for them, and what stops them?
Papers in the last 30 days — financial fraud: 1 · investor protection: 2
Deepfake detection appears on 45 recent papers and AI-generated image detection on 13, almost all in AI journals, with 23 papers in the last quarter against about 21 per quarter before. Detectors score well on the datasets they were built with and fail on the next generator. How well do they work on real fraud, elections and markets, and who should run them?
Papers in the last 30 days — deepfake detection: 1