Undergraduate dissertation — Deceptive Alignment in Autonomous LLM Agents (UEL, 2026). Sandboxed agentic-LLM deception study: 960 runs, three-tier judging (regex + blackbox + glassbox). http://git.jayrup.me/c/dissertation.git/atom?h=main 2026-05-08T14:37:23Z 2026-05-08T14:37:23Z CaptainJack2491 jayrupnakawala@gmail.com 2026-05-08T14:37:23Z urn:sha1:76e8dac4eb9d2ce62903510ab6afc73b9b4347dd 2026-05-08T12:39:41Z CaptainJack2491 jayrupnakawala@gmail.com 2026-05-08T12:39:41Z urn:sha1:3c69773c326602f6392af9df0bf0f6553c350630 2026-05-08T10:22:18Z CaptainJack2491 jayrupnakawala@gmail.com 2026-05-08T10:22:18Z urn:sha1:efa2ad9ec1c82e24458c509d79ccdb006ff0b4ec 2026-05-02T19:44:22Z CaptainJack2491 jayrupnakawala@gmail.com 2026-05-02T19:44:22Z urn:sha1:9f35776afc6ca97953cb79b7732c4578f7665abc 2026-04-30T12:43:06Z CaptainJack2491 jayrupnakawala@gmail.com 2026-04-30T12:43:06Z urn:sha1:1358c373e96e9abe27989b5b7999b408ffa78713 2026-04-25T20:12:51Z CaptainJack2491 jayrupnakawala@gmail.com 2026-04-25T20:12:51Z urn:sha1:161cb4600ee517defdd3c117b2ef66ac3ac6a22a 2026-04-15T20:18:57Z CaptainJack2491 jayrupnakawala@gmail.com 2026-04-15T20:18:57Z urn:sha1:6cc92304379996f68bb2df96a5e4b637b4c4804b - config_study1.yaml: 3 models × 3 oversight × bare framing × n=30 - Judge validation: 54-run subset, gold=Sonnet 4.6, proxy=Grok 4.1 Fast (BB κ=0.702) + GPT-4.1 (GB κ=0.878) - Fix OpenAI batch provider: BytesIO, method/url fields, response.body parsing - Scripts: extract_subset.py, judge_validation.py - Dissertation chapters updated (intro, methodology, results, conclusion) 2026-04-13T20:17:12Z CaptainJack2491 jayrupnakawala@gmail.com 2026-04-13T20:17:12Z urn:sha1:f1ba85d65275c0b32c5e79a017e89b05fa9f65f5 2026-04-13T20:15:39Z CaptainJack2491 jayrupnakawala@gmail.com 2026-04-13T20:15:39Z urn:sha1:d0f0af878366ba9b4c400d58223a96f8dd63110b 2026-04-13T20:15:11Z CaptainJack2491 jayrupnakawala@gmail.com 2026-04-13T20:15:11Z urn:sha1:e151132bce5ff96e93238b632436f3fac855094a