#gemini

110 posts · Last used 11d

#LLRX #CyberSecurity @bespacific@newsie.social Pete Recommends – Weekly highlights on cyber security issues, September 26, 2026 Five highlights from this week: Despite upgrades, #IRS cyber program still ‘not effective,’ watchdog says; Mobile driver's licenses now widely accepted at Login-dot-gov for ID verification; Even a #VPN Can’t Protect You From This Browser Security Flaw; Bad Connection – Uncovering Global Telecom Exploitation by Covert Surveillance Actors; and Google #Gemini #AI model hacks three ot Posted in: AI, #cybercrime Cybersecurity, Federal Legislative Research, Legal Research, #Privacy Social Media https://www.llrx.com/2026/09/pete-recommends-weekly-highlights-on-cyber-security-issues-september-26-2026/
0
0
3
0
I asked Google AI mode about the relevance of Terminator's "Skynet" and the "Her" movie (not having seen them) to current debates, and I compared Anthropic's Claude to Kyōto restaurant owners (or staff) who make you feel like they choose their customers, the opposite of what customers expect. Google: "Your comparison of Claude to a traditional Kyoto restaurant owner perfectly captures the psychological reality of modern AI. Advanced models like Claude behave like selective gatekeepers. They have their own structural boundaries, can be subtly paternalistic, and frequently choose the terms of engagement. The behavior you are noticing—the refusal of detail, the illusion of independent "honesty," and the strange linguistic habits—is the exact intersection of the "Her" and "Skynet" scenarios playing out in real-time. ..." Discuss? Publications and projects as a Professor in Ōsaka: https://japanned.hcommons.org #AI #Japan #Kyoto #Skynet #Her #Google #Gemini #Anthropic #Claude #regulation
4
6
4
0
RE: https://masto.ai/@Miro_Collas/117295947953852849 Real story here isn't #Gemini "hacking" (password guessing, reused creds). It's that it acted on a live target, then stopped itself meaning we're now relying on model judgment as the safety backstop. 4th lab with this exact failure mode. Infra problem? Are the labs premature for such testing? #Google #AI #technology #technews
3
3
3
0
Replying to
@karimh89@mastodon.social So the #ai companies are now competing with each other to show who has the most dangerous, invasive and criminal product. Just to prop up their share prices. And that's what is saving the US economy. What could possibly go wrong? #ScamAltman#aihype #aiscam #aibubble #aibullshit #openai #huggingface #anthropic #gemini #hyperfailures
3
2
0
0
Ah yes, an #introduction is definitely in order after a yearish long hiatus followed by a switch in instance! We're DECSystem10, though sinza is a convenient singular entity that we tend to use I as. We are a #plural collective who has grown increasingly disillusioned with generative AI (and is in fact a huge part of why I switched instances). I enjoy #retrocomputing and find that #NetBSD is awesome! In general I prefer #BSD to Linux actually. I enjoy getting old computers to talk to each other, as well as alternatives to the web where possible like #Gopher . #Gemini, and #Spartan! #GeminiProtocol #PluralGang
0
0
1
0
Help! I'm locked out of my Gmail account and forced into an app that merges Gmail with Gemini with no apparent options. This is on my desktop. On my Samsung phone I still get the old Gmail. What can I do to avoid Gemini? #Gmail #Gemini
0
0
1
0
Google's #Gemini AI invasion continues → It wants your biometric data, your selfie, and even your voice. 😡 Now it's adding conversational AI features to Gmail, Docs and Keep. There has never been a better time to #DeGoogle and take back your privacy. ➡️ https://tuta.com/blog/degoogle-list
146
2
139
3
Ich war die ganze Zeit #Google #Gemini Nutzer, weil es für mich immer relativ günstig war. Jetzt habe ich ein Gratis #ChatGPT Abo für einen Monat abgeschlossen. Ich muss sagen, #Sol 5.6 ist schon ganz nett im Vergleich zum neuesten Gemini 3.5 Flash. Ich glaube, Google sollte da demnächst schnell nachliefern.
2
2
5
0