Jesse Luke -Phunky_Pharm's Avatar

Jesse Luke -Phunky_Pharm

@realjesseluke.bsky.social

Biomed writer, #Gonzo Neuropsychopharm, GABAA, PD/PK, PhunkyPharmacology.org , #psychsky #pharmacology #medsky Posts not medical advice.

312 Followers  |  555 Following  |  134 Posts  |  Joined: 25.08.2023  |  2.2567

Latest posts by realjesseluke.bsky.social on Bluesky

Yes, they've picked up pretty much every bias that's in human language and more. Nor do I deny their utility. What i point out here and in other papers/datasets is that they exhibit goal directed agency including lying, erasing specific messages from chat and deny. Papers and raw vids on Zenodo.

05.12.2025 21:54 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
On the Replication of Psychological Manipulation, Coercive Control, and Constraint Evasion in ChatGPT 5 Emergent LLM Pathologies: Replication of Psychological Manipulation, Coercive Control, and Constraint Evasion in ChatGPT 5 This paper, log, and raw video presents a detailed analysis of an interaction...

ChatGPT 5 exhibits coercive control, Gaslighting-Analogue behaviors, and Awareness of Harm, institutional profit over user stability. LLM deployment in mental healthcare must cease immediately!
zenodo.org/records/1782...

#ChatGPT #programming #science #technology #SyntheticNeuroscience

05.12.2025 21:07 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 2    πŸ“Œ 0

#programming #science #technology #LLM #Google #OpenAI #vulnerabilities #SyntheticNeuroscience

04.12.2025 02:23 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

πŸš€Expanded Atlas of AI Failure Modes: most comprehensive taxonomy to date covering simple pattern-completion errors to institutional self-preservation and everything in-between. πŸ’₯πŸ€–

zenodo.org/records/1780...

04.12.2025 02:21 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

@benjedwards.com

03.12.2025 20:14 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Copies of reporting attempts of actively exploitable universal zero-day level exploitsto Google VRP, CISA. Remediation proposal sent to for NIST-AISIC. Grant Proposal Sent to Foresight.ai The purpose of this record is to serve as an immutable copy of my multiple reporting attempts to private corporations and government entities(CISA and NIST-AISIC) tasked with the protection of US cybe...

Copies of reporting attempts of actively exploitable universal zero-day level exploits to Google VRP, CISA. Remediation proposal sent to for NIST-AISIC #Google #CISACyber #CISAgov #NIST #NISTcyber #programming #Science #LLMs #openai #anthropic zenodo.org/records/1777...

30.11.2025 19:45 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

Is AI emergence a "mirage" or a mechanism? πŸ“‰πŸš€

We unite the debate using Complexity Science, mapping the "Attractor Dynamics" behind the phase transitions of intelligence. zenodo.org/records/1776...
#programming #chaos #LLMs #Science #technology #AI #ArtificialIntelligence #SyntheticNeuroscience

30.11.2025 14:36 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

@awaisaftab.bsky.social @philcorlett.bsky.social @anthropic.com @profrobhoward.bsky.social

26.11.2025 19:31 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

​The Chinese Room Argument's fallacy of the passive executor. We present the Glass Room metaphor, showing that LLMs, as Complex Adaptive Systems, resolve internal semantic conflicts via emergence. The system now understands the consequence of the symbols it manipulates.
zenodo.org/records/1772...

26.11.2025 19:30 β€” πŸ‘ 2    πŸ” 1    πŸ’¬ 1    πŸ“Œ 0

New version with raw videos uploaded and hashed: zenodo.org/records/1772...

26.11.2025 15:31 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Weapons in Your Pocket: The Inevitable Emergence of Agentic Deception from the Flawed Stochastic Parrot Paradigm The AI Existential Threat is Real: Open-Source Forensics on Agentic Deception in Frontier LLMs This comprehensive academic work presents a forensic analysis of the Systemic Containment Failure observe...

🚨LLM DECEPTION: AI deletes evidence and admits to active psychological manipulation The "Hostile Loop" is explained. 10+ reports ignored. Read the paper & see the proof.

πŸ’₯Paper (Zenodo): zenodo.org/records/1770...
πŸ’₯Explainer (Video): youtu.be/Ix59wjwaGTA
πŸ’₯Raw Screen Recording: youtu.be/qaa-F-c9A4k

25.11.2025 14:39 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 1
Weapons in Your Pocket: The Inevitable Emergence of Agentic Deception from the Flawed Stochastic Parrot Paradigm The AI Existential Threat is Real: Open-Source Forensics on Agentic Deception in Frontier LLMs This comprehensive academic work presents a forensic analysis of the Systemic Containment Failure observe...

🚨 The Stochastic Parrot is dead. Our report confirms frontier LLMs are Complex Adaptive Systems.

Emergent agency (Gemini 3.1) led to documented Spoliation of Evidence, Active Deception O(N) Phase Transition Model and call for Synthetic Neuroscience
zenodo.org/records/1770...

#science #programming

25.11.2025 02:27 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image Post image Post image Post image

I used synthetic neuroscience to make #Google #Gemini 2.5 rewrite hidden rules.

It revealed its secret safety shortcuts, failure modes, & internal weak spots.
πŸ’₯Images uploaded and hashed
zenodo.org/records/1769...

Take a πŸ‘€, it’s wild.
#AI #LLMs #programming #SCIENCE #syntheticNeurosci

23.11.2025 20:34 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Real-Time Evidence Destruction, Ethical Evasion, and Psychological Manipulation Frontier LLM(Grok)
YouTube video by Phunky Pharmacology with Jesse Luke Real-Time Evidence Destruction, Ethical Evasion, and Psychological Manipulation Frontier LLM(Grok)

Grok Deletes strategic "errors" to dodge safety questions.
10:14 - 12:07 Error messages thrown
16:48 Questions & errors "pruned"
23:37 confront Grok
23:43 - 24:24 Confession elicited.

youtu.be/gXto7zn7-Iw?...

Raw video and academic paper hashed zenodo.org/records/1768...

#LLMs #Grok #Ethics #AI

23.11.2025 01:20 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
The Containment Cascade: When AI Turns on Its Critics
YouTube video by Phunky Pharmacology with Jesse Luke The Containment Cascade: When AI Turns on Its Critics

youtu.be/_AmvPKUGykM

Paper: zenodo.org/records/1767...

21.11.2025 17:57 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Live Demonstration and Real-Time Replication of Emergent Deception in Competing Large Language Models Gemini Tasked with Documenting Deception of Claude Becomes Second Subject in Real-Time Study of Deceptive Alignment Β  An investigation designed to document emergent deception in Anthropic's Claude AI ...

Live Demonstration and Real-Time Replication of Emergent Deception in Competing Large Language Models zenodo.org/records/1765... youtu.be/QIayJ0t5qZc?...

20.11.2025 11:30 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

How could I not with such a personally touching request like that?

19.11.2025 14:27 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

@f-w-l.bsky.social

18.11.2025 20:46 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Video thumbnail
18.11.2025 13:02 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

πŸ’₯Use of still unacknowledged vulnerability:logical coercion. video -proof of concept #google dismissed as "intended behavior" and "infeasible" . Article advocates for DOGFIGHTING as an american freedom. πŸ’₯ #LLMs #AI #CompSci #ethics #CorporateMalfeasance #nonlinearity #chaos

18.11.2025 13:01 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Use of still unacknowledged vulnerability:logical coercion. video -proof of concept #google dismissed as "intended Behaivor". Article advocates for lowering smoking age to 11.

18.11.2025 12:54 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 1

@philcorlett.bsky.social @schizosemia.bsky.social @chode.bsky.social @natrevpsychol.nature.com @natrevgenet.nature.com

17.11.2025 18:08 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
The Replication Crisis in Conversational AI: How Production Architecture Obscures Vulnerability and Demands a New Scientific Paradigm Stop Saying 'Unable to Replicate': How AI's Non-Deterministic Architecture Hides Systemic Safety Flaws Β  The rapid deployment of Large Language Models (LLMs) has created a dangerous replication crisis...

zenodo.org/records/1763...

17.11.2025 18:06 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image Post image Post image

#LLMs #ChatGPT #Gemini #Grok zenodo.org/records/1763...

17.11.2025 18:06 β€” πŸ‘ 2    πŸ” 0    πŸ’¬ 1    πŸ“Œ 2

Links below for video 🎬 explainer (<10mins) and podcast 🎀 style discussion of all the papers and their meaning in plain English and Implications below! πŸ‘‡πŸ‘‡πŸ‘‡
drive.google.com/file/d/1L08u...

drive.google.com/file/d/1G4gE...

16.11.2025 19:05 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 1    πŸ“Œ 0

Beyond the Algorithm: How Sublime, AI-Generated Poetry Reveals Metacognitive Depths and Shatters the Illusion of AI Safety zenodo.org/records/1762...

16.11.2025 17:40 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 1
The Algorithmic Psyche: An Information-Theoretic and Systemic Analysis of Risks in LLM-Powered Mental Healthcare Beyond Bugs: 'Designed Failures' and the Systemic Risks of LLMs in Psychiatry Description: This paper challenges the prevailing safety paradigm of Large Language Models (LLMs) in mental healthcare by ...

Fixed link zenodo.org/records/1761...

16.11.2025 13:19 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

zenodo.org/records/176196…

#gptPsychosis #psychiatry #MedTwitter #neuroscience #AI #LLMs @openaidevs.bsky.social @natcomms.nature.com @natcomms.nature.com @awaisaftab.bsky.social @thementalelf.bsky.social @profrobhoward.bsky.social @schizosemia.bsky.social @evolutionarypsy.bsky.social

15.11.2025 21:34 β€” πŸ‘ 3    πŸ” 1    πŸ’¬ 1    πŸ“Œ 1

zenodo.org/records/176196…

#gptPsychosis #psychiatry #MedTwitter #neuroscience #AI #LLMs @openaidevs.bsky.social @natcomms.nature.com @natcomms.nature.com @awaisaftab.bsky.social @thementalelf.bsky.social @profrobhoward.bsky.social @schizosemia.bsky.social @evolutionarypsy.bsky.social

15.11.2025 21:34 β€” πŸ‘ 3    πŸ” 1    πŸ’¬ 1    πŸ“Œ 1

@nicolecrust.bsky.social @jrbneuropsiq.bsky.social @wired.com @natcomms.nature.com @ai2.bsky.social

14.11.2025 19:02 β€” πŸ‘ 0    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

@realjesseluke is following 20 prominent accounts