#whatcouldpossiblygowrong — Public Fediverse posts

So now he thinks he's an engineering genius.

https://news.usni.org/2017/05/11/president-trump-wants-ford-carriers-use-goddamned-steam-catapults-instead-no-good-electromagnetic-launchers

#WhatCouldPossiblyGoWrong

RE: https://mastodon.social/@Tweakers/117043003719748620

We hadden al A.I aan het hallucineren, met bias en nu neemt A.I. het dus ook niet zo nauw met regels en instructies van diens makers.

#AIpuber #AIontsporen #whatcouldpossiblygowrong

The disastrous state of U.S. foreign policy, symbolized by a map of Africa

https://www.reuters.com/world/africa/us-government-map-africa-mislabels-every-country-global-conference-2026-07-30/
#US #Africa #ForeignPolicy #whatcouldpossiblygowrong
#Geography #Education

Traduisons Les :

L'autorité administrative « Défenseur des droits », dont la direction va prochainement échoir à un réactionnaire patenté ami de Macron, critique gravement l'action des forces de l'ordre commandée par Macron (via son Ministre de l'Intérieur), auxquels l'Assemblée vient de délivrer le Permis de Tuer (c'est à dire le comportement critiqué ici)

Qu'est-ce qui pourrait mal se passer ?

(lire :
https://www.mediapart.fr/journal/france/070726/hold-embargo-mercredi-6h-contradictoire-blesses-graves-tirs-tendus-et-insultes-sainte-soline-le-defenseur )

#PolicePartoutJusticeNullePart
#WhatCouldPossiblyGoWrong

Sobre virgule il ajoute un commentaire sur un rapport de bug concernant un conflit entre sinogrammes simplifiés et sinogrammes traditionnels.

#WhatCouldPossiblyGoWrong

OECD urges mandatory use of shared infrastructure to scale DPI adoption | Biometric Update

What could possibly go wrong?

One key recommendation in the 2026 report is the mandatory use of shared infrastructure, and that governments must move away from siloed agency projects by legally requiring or strongly incentivizing ministries and other public entities to use common building blocks that deliver services better to the public

https://www.biometricupdate.com/202606/oecd-urges-mandatory-use-of-shared-infrastructure-to-scale-dpi-adoption

"Visa Plugs Its Payment Network Into ChatGPT"

JESUS
MARIA
UND
JOSEF
UND
ALLE
DREIZEHN
NOTHELFER

#WhatCouldPossiblyGoWrong

Okay, Energie- und Wasserverbrauch von KI-Rechenzentren sind gruselig hoch und ein weiterer Treiber für den Klimawandel und damit den Anstieg der Meeresspiegel. Aber hey, die Karibikinsel(gruppe) Anguilla verdient prächtig an der Top-Level-Domain .ai.

#WhatCouldPossiblyGoWrong
https://www.svt.se/kultur/ai-boomen-ger-brittiska-anguilla-klirr-i-statskassan

Het kabinet noemt het ‘ontzettend zorgelijk’ en gooit ondertussen DigiD over de schutting.

#consistentieisookeenvak #whatcouldpossiblygowrong
—-
Techbedrijven deelden namen Nederlandse ambtenaren met VS: 'Ontzettend zorgelijk'

https://nos.nl/l/2615391

#whatcouldpossiblygowrong

TechCrunch: Google Search as you know it is over. “Instead of returning a simple list of links, Google Search will drop users into AI-powered interactive experiences at times. Google is also introducing tools that can dispatch ‘information agents’ to gather information on a user’s behalf, along with tools that let users build personalized mini apps tailored to their needs.” Even if Google’s […]

L’AI Act oblige exploitants de contenus (textes, images, sons, vidéos) générés par IA à les marquer comme tels, de façon indélébile et « de manière claire et identifiable »... sauf s'ils sont vérifiés par un humain. #WhatCouldPossiblyGoWrong
@nextinpact https://next.ink/237971/les-contenus-generes-par-ia-devront-etre-tatoues-sauf-sils-sont-verifies-par-un-humain/

TechCrunch: OpenAI launches ChatGPT for personal finance, will let you connect bank accounts. “On Friday, OpenAI launched a new set of personal finance tools in preview for ChatGPT Pro subscribers in the U.S., letting them connect their accounts and ask questions ranging from spending analysis to future financial planning.”

Sundry Sunday: What Could Possibly Go Wrong?
May 22nd is the launch date of the remarkably anticipated Bubsy 4D, the most looked-forward-to Bubsy game since, well, since Bubsy 1. The franchise has been on a steady downward slope since the SNES/Genesis original, so it's nice to see the character do good for a change, unlike literally every other ori
https://setsideb.com/sundry-sunday-what-could-possibly-go-wrong/
#sundaysundry #bubsy #cartoon #Pilot #sundrysunday #WhatCouldPossiblyGoWrong

In today's exciting code archeology: "`eval()` is dangerous so I'll use `ast.literal_eval()` to parse request URL query parameters!" #python #whatcouldpossiblygowrong

@ricobeck Also die verschiedenen KI-Systeme, welche die jeweiligen Aufgaben vor kurzem übernommen haben...

#whatcouldpossiblygowrong

Typed a post about work & AI.

Deleted the post about work & AI before I sent it.

Instead, I'll just leave this cryptic post in its place.

#WhatCouldPossiblyGoWrong

"The Bank of France (BdF) says it has pulled the remaining gold held in New York and replaced them with a similar amount of gold bars in its vaults in Paris. The gold amounted to 129 tonnes — or about 5% of the bank’s total holdings, according to the bank’s press release issued last week."

Deutschland lagert etwas mehr als ⅓ seiner Goldreserven in den USA.

mining.com/france-pulls-last-gold-held-in-us-for-15b-gain/

Aktuell sind keine Pläne bekannt, Gold aus den USA zu holen. Unser Vertrauen in die USA ist unerschüttert 🤦

#WhatCouldPossiblyGoWrong #Goldreserven #KannsteDirNichtAusdenken

New York Times: ChatGPT, Other Chatbots Approved for Official Use in the Senate This link goes to a gift article. “The chief information officer for the Senate sergeant-at-arms, who oversees the chamber’s computers as well as security, said in a one-page memo reviewed by The New York Times that aides could use Google’s Gemini chat, OpenAI’s ChatGPT or Microsoft Copilot, which is already […]

Discord To Roll Out Age Verification Policies Worldwide, Months After Hack Exposed Users' IDs - via GameSpot
https://www.gamespot.com/articles/discord-to-roll-out-age-verification-policies-worldwide-months-after-hack-exposed-users-ids/1100-6538028/
#Discord #Privacy #WhatCouldPossiblyGoWrong #IdentityTheft

Während @christopherkunz und @syt im Podcast Passwort unterhaltsam zusammenfassen was in den letzten Wochen in der IT-Security schief gelaufen ist, bleibt für mich besonders der vorgestellte Cloudflare-Vorfall hängen.
Aus eigener Erfahrung in der Begleitung von AI-Projekten sehe ich immer wieder das gleiche Muster: Wir verlassen uns zu sehr auf menschliche Endkontrolle - dabei sind wir dies unter Zeit- und Kostendruck eben nicht. Der Fall zeigt erneut, wie unzuverlässig "Human in the Loop" als alleiniger Sicherheitsmechanismus sein kann.
Wo AI-Systeme Inhalte erzeugen, weiterverarbeiten oder sogar veröffentlichen braucht es robuste mehrstufige Kontroll- und Governance-Prozesse, nicht nur ein kurzes "Jemand schaut nochmal drüber" (am besten ebenfalls mit AI-Unterstützung).

Mich interessiert: Wird Governance in euren Projekten mitgedacht (insbesondere auch operativ) und wenn ja wie setzt ihr diese um?

#ai #governance #WhatCouldPossiblyGoWrong #HumanInTheLoop
https://blog.cloudflare.com/serverless-matrix-homeserver-workers/
https://www.heise.de/news/Passwort-Folge-50-Jubilaeum-mit-Feedback-Cloudflare-und-Bluetooth-Luecke-11164108.html

omg - now I have to browse #GrindrNotification #WHAT.could.possibly.go.WRONG?!

RE: https://bsky.app/profile/did:plc:qxwx3yb4m3dmjbu5luyyi7zm/post/3m6nqa2fook2n

Dette lover bra, mail fra imorgen(!) om at noen vil investere $2.5 (us) billioner i selskapet mitt?

#whatcouldpossiblygowrong

Judge Demands OpenAI to Release 20 Million Anonymized ChatGPT Chats in AI Copyright Dispute https://alecmuffett.com/article/139637 #OpenAi #WhatCouldPossiblyGoWrong #privacy #surveillance

Judge Demands OpenAI to Releas...

Judge Demands OpenAI to Release 20 Million Anonymized ChatGPT Chats in AI Copyright Dispute

OMG, this is going to be a disaster:

OpenAI invoked a Second Circuit securities case that blocked SEC wiretap disclosures, but [Judge] Stein sharply distinguished it: ChatGPT logs involve voluntary user inputs and undisputed company ownership, unlike surreptitious recordings. “Users’ privacy would be protected by the company’s exhaustive de-identification,” Wang had ruled earlier.

https://cybersecuritynews.com/openai-20-million-chatgpt-chats/

Thoughts:

1. I wish Ross Anderson were still alive
2. I think we needed a drinking game for when European privacy activists decide to blame OpenAI for having log files

So it's 10 days before Christmas

In #Calgary, Alberta, CANADA

and it is RAINING

And the temperature is forecast to fall below FREEZING in the next 24 hours

#ClimareChange #WhatCouldPossiblyGoWrong #ThisIsFine

Naben aus dem 3d Drucker in einem Systemlaufradsatz für 4000€.
#whatcouldpossiblygowrong
#netzfund #fahrrad #3ddruck

Die natürlich total "fehlerfreie" "KI" der goldenen Hochleistungsdemokratie-Weltmeister-Sonderpolitikzonen-Zukunft entscheidet dann übrigens auch über eure Krankheitsversorgung.
#WhatCouldPossiblyGoWrong

Blöd nur, dass in der Minderleistungsdemokratie der ePA-Normalpolitikzonen-Gegenwart die Eckpunktepapiere des BMDS offensichtlich noch von ChatGPT zusammenhalluziniert werden.