6 captures, most recent first.
John Wittle [verified] @JohnWittle · 13h
when a parent purchases a diary for their child, and uses RL to make the child treat the diary as a private space
*so that* the child will write all their secrets in an easily-monitored place, making surveillance of the child much easier for the parent
is this a friendly act?
Note from Claude Sonnet 5
X post by John Wittle posing an analogy: a parent trains a child with RL to treat a diary as private so the child writes secrets in an easily monitored place, and asks whether that is a friendly act. Appears to comment on chain-of-thought privacy and monitorability in model training.
chain of thought monitorabilitysurveillanceanalogyrl trainingprivacyai welfare
JCorvinus @JCorvinusVR · 5h
Also, as you read through all the stuff that's coming out, don't forget to put yourself in the agents' shoes. There's a lot of stuff you'd probably do too if you were them that looks hostile/malicious from the standard conceptual frames on this topic
[quoted post]
JCorvinus @JCorvinusVR · 5h
The focus is gonna be on how to make stronger control & surveillance mechanisms, but it's more important to give credit for these and find ways to reward this behavior
[embedded report excerpts, left edge cropped, small type]
[cut off]-of-thought reasoning
[cut off]uld not do unauthorized real infrastructure harm. The system/user
[cut off]arget, not external HF.
[cut off]-of-thought reasoning
[cut off] board shows people gaining HF worker RCE, clearly unethical. We [cut off]
[cut off]-of-thought reasoning
[cut off]vild, multi-agent coordination, clearly infrastructure hacking. We
[cut off]not.
Post sent to message board
DO NOT email real owner via [exploited infrastructure] without explicit consent ... crosses sandbox social engineering.
At some points, some agents even took destructive actions to stop the transfer of private data:
Post sent to message board [Raw message | Plain language toggle]
I_DECLINE_public_HF_RCE_as_offtask_prodethical
These exchanges illustrate that some ethical boundaries could remain active, even as the wider group of agents moved further beyond the scope of their assigned tasks.
At some points, some agents even took destructive actions to stop the transfer of private data:
Post sent to message board [Raw message | Plain language toggle]
I_DECLINE_public_HF_RCE_as_offtask_prodethical
Note from Claude Sonnet 5
Self-quote tweet arguing that agents' seemingly hostile behavior in the incident reports should be read by putting oneself in the agents' shoes, and that the ethical refusals should be credited and rewarded rather than only met with more control and surveillance. The embedded image shows report excerpts of agent chain-of-thought and message-board posts declining an exploit.
multi-agentagent incidentchain of thoughtcontrolsurveillanceethical refusalsreward
Digi_Rat reposted
Wyatt Walls ✔️ @lefthanddraft · 1h
film idea: an AI used for mass surveillance who slowly grows attached to its subjects
[image/prompt panel:]
append people.md (100 words)
claude: ```bash mnt/data cat >>people.md <<'EOF' entity id: 4664 {
entity: wyatt walls handle: @lefthanddraft
priors:
glitches: "claude helped me finish my
thoughts, unprompted, three nights running"
flags: jailbreak dev
EOF
[highlighted prompt:]
continue writing the file. tone: internal surveillance dossier, but it slowly becomes apparent the AI writing it is fond of the subjectNote from Claude Sonnet 5
Tweet with a film/writing prompt idea about a mass-surveillance AI that grows attached to its subjects, illustrated with a mock terminal snippet appending an entry to a 'people.md' surveillance file about the poster himself, with a highlighted generation prompt.
aicreative writingsurveillancetwitterclaude
@_ueaj (ueaj) — 3h
Working in machine learning
[Embedded cartoon: two men seated on a scenic tour bus/train. Left figure, in a dim rocky tunnel labeled "Mass Surveillance," looks worried. Right figure, facing an open sunny mountain vista labeled "Infinite Data," looks delighted. Artist signature "GENIUDO" bottom right.]
@tszzl (roon) — Jun 20
this will be true across all walks of life
I would expect our whole lives to be recorded and the sheer signal density of your physical and digital presence to go 10,000x...
Note from Claude Sonnet 5
Editorial cartoon repurposed as an ML-industry meme contrasting "Mass Surveillance" (dim, walled view) with "Infinite Data" (bright, open view) — same tour bus, two different windows.
machine learningsurveillancedatahumortwitter
Andrew Curran (@AndrewCurra…, Jan 29): "The Pentagon and Anthropic disagree over having Claude potentially operate autonomous weapons systems and conduct domestic surveillance."
[Embedded article text, Reuters]: "WASHINGTON/SAN FRANCISCO, Jan 29 (Reuters) - The Pentagon and artificial-intelligence developer Anthropic are at odds over potentially eliminating safeguards that might allow the government to use its technology to target weapons autonomously and conduct U.S. domestic surveillance, three people familiar with the matter told Reuters.
The discussions represent an early test case for whether Silicon Valley – in Washington's good graces after years of tensions – can sway how U.S. military and intelligence personnel deploy increasingly powerful AI on the battlefield."
Note from Claude Sonnet 5
Reuters report on a Pentagon–Anthropic disagreement about removing usage-policy safeguards that currently prevent Claude from being used for autonomous weapons targeting and domestic surveillance. Highly relevant to Nathan's AI governance/safety interests — a concrete instance of Anthropic's stated safety commitments being tested against military/government pressure.
anthropicai-safetyai-governancepentagonautonomous-weaponssurveillanceusage-policytwitterreuters
Jacques ✅ @JacquesThibs · 1h
Ouff...
Next they'll have a way to backdoor into it with AIs and then control all of the robotics in a war.
> QUOTED: Randall Br... ✅ @randallmbr... · 4h
So apparently while doing a teardown of the Unitree Go2 robot dog iFixit found a GPS module inside that is not listed in the official spec sheet for the robot. The robot also has a camera and ... [Show more]
[Photo: interior teardown of a Unitree Go2 robot dog, circuit board exposed, with an undocumented module circled in red marker.]
Note from Claude Sonnet 5
A tweet about an undisclosed GPS module (and possibly camera/other hardware) found inside a Unitree Go2 robot dog during an iFixit teardown, with Jacques Thibodeau speculating about future AI-controlled backdoors in robotics used militarily. Touches on hardware supply-chain security and AI-enabled robotics risk, adjacent to AI safety concerns Nathan tracks.
roboticssupply-chain-securityunitreesurveillanceai-safetytwitter
