@hex@kolektiva.social2025-10-30 19:36:00
As a security engineer, whenever anyone talks about a control it's always important to ask "by what mechanism?"
> "Oh, that can't happen because we have a system to stop it."
By what mechanism?
> "We have documentation that says...."
Yeah, that's not a mechanism.
People keep saying, "Trump can't do that!" But like... by what mechanism?
> "The constitution says..."
Yeah... a documented list of rules is not self-enforcing. What is the mechanism?
What makes this impossible? Oh, it's possible under certain conditions? Oh, it's always possible and you're completely relying on the idea that there will never be a malicious actor? Yeah, that's gonna get exploited. Oh shit, now you're owned.
What do you do with a system that's completely owned? Once it's compromised it can never be trusted again. What would you tell a client who told you, "Patching is really hard, so we're just gonna ban the attacker's IP."
What, you're not even gonna reinstall?
I assume we've all had the "burn everything down and start again" client. I wonder how many of us thought we would see the US government ask for them to hold it's beer.
#USPol
@heiseonline@social.heise.de2025-11-30 06:34:00
Missing Link: ChatGPT als Therapeut – Die Wirkung der Resonanzmaschine
Immer mehr Menschen vertrauen einer Maschine, die nichts fühlt, aber uns immer besser versteht. Welche Folgen hat es, wenn ChatGPT therapeutisch wirkt?
@seeingwithsound@mas.to2025-11-30 08:55:24
Non-invasive technology can shape the brain's reward-seeking mechanisms https://www.plymouth.ac.uk/news/non-invasive-technology-can-shape-the-brains-reward-seeking-mechanisms Non-invasive ultrasonic neuromodulation of the human …
@pavelasamsonov@mastodon.social2025-09-29 13:13:48
AI didn't create skewed workplace incentives. They were already aligned towards churning out artifacts to create the illusion of productivity.
Of course, if there is one thing AI is good at, it's making plausible-looking artifacts. Then those artifacts get passed on and become your problem.
HBR is calling this the workslop era.
We must kill the logic underlying workslop: that anything going wrong after a handoff can be blamed on execution problems.
@cdarwin@c.im2025-12-01 00:29:13
The assumption is that one day large language models and other related AI technologies fostered by Google Gemini and OpenAI ChatGPT actually will be a great and infallible productivity tool for genuine work.
It already is decent for providing basic overviews of highly-covered, well-sourced topics, even as hallucinations and sycophancy continue to dog the tech, particularly in situations where accountability is more critical.
Despite today's downsides,
many companies ar…
@memeorandum@universeodon.com2025-10-30 13:05:56
Federal Workers Are Barely Making It Through the Government Shutdown (Wired)
https://www.wired.com/story/federal-workers-barely-making-it-through-government-shutdown/
http://www.memeorandum.com/251030/p38#a251030p38
@primonatura@mstdn.social2025-10-30 12:00:15
"Organised crime making millions from illegal waste dumping in UK, says committee"
#UK #UnitedKingdom #Environment
@heiseonline@social.heise.de2025-10-01 05:27:00
Chefin der Signal-App droht mit Rückzug aus Europa
Signal gilt als einer der sichersten Messenger. Die App könnte aber in Europa vom Markt verschwinden, falls sich die Chatkontrolle in der EU durchsetzt.
https://www.
@memeorandum@universeodon.com2025-11-30 14:55:24
OpenAI is a loss-making machine, with estimates that it has no road to profitability by 2030 -- and will need a further $207 billion in funding even if it gets there (Jez Corden/Windows Central)
https://www.windowscentral.com/artificial-intelligence/openai-chatgpt/analysis-openai-is-a-loss-making-machine
http://www.memeorandum.com/251130/p9#a251130p9
@heiseonline@social.heise.de2025-09-30 10:02:00