@anecstandards.bsky.socialThe CEN-CENELEC JTC 21 Plenary starts today in Winterthur, Switzerland!
ANEC representatives are there to discuss the latest developments in #AI standardisation and ensure that consumer interests remain at the heart of European AI standards.
@infor-pl.bsky.socialFaktyczny nadzór człowieka nad algorytmem jako wymóg prawny ⤵ #aiact #rekrutacja
kadry.infor.pl/kadry/hrm/re...
@pcmaggreece.bsky.social
@netopia.bsky.socialOpenAI to add some watermarking by way of self-regulation in the EU to comply with #AIAct
Pro-tip: Marking one's own homework is never a good idea and has failed multiple times, entrenching the infringement of IP
@justgeekfr.bsky.social
@denyally.bsky.socialChatGPT text gets an EU watermark
OpenAI says it will watermark ChatGPT and Codex text in the EU under the AI Act. Editing may make the marks harder to detect—because apparently even invisible ink needs…
via TechCrunch AI
#OpenAI #ChatGPT #AI #AIAct #gamedev #indiegame #indiedev #indiegames
@news-it.bsky.socialOpenAI、テキスト用の電子透かし「textGrain」を導入 ChatGPTとCodexはEUから
#textGrain #AIAct #ITニュース
@today0tech.bsky.socialOpenAI will watermark ChatGPT and Codex text in the EU to comply with the AI Act, noting that editing can make the invisible marks harder to detect.
#OpenAI #ChatGPT #AIAct #EUAi
https://techcrunch.com/2026/10/05/openai-will-start-watermarking-chatgpts-text-in-the-eu/
@winfuture.de#OpenAI führt in der EU unsichtbare Wasserzeichen für #ChatGPT ein, um den #AIAct zu erfüllen. Ein statistisches Muster in der Wortwahl verrät die Texte.
@yuuiko7.bsky.socialOpenAI is rolling out text watermarking for ChatGPT and Codex outputs in the EU to comply with the AI Act
- EU users get watermarking by default over the coming weeks
- API customers globally can opt in for select models
- Covers both chat and code generation (Codex)
@knoware-project.bsky.social📬 Our first newsletter is here!
Inside: our kick-off in Trondheim, the launch of our website, and a new series, 3+1 Facts about AI.
Read it here 👉https://tinyurl.com/yckdvsc8
#AIAct #ResponsibleAI #AISafety #GPAIGovernance #AICapabilities #HorizonEU #ResearchImpactEU #HaDEA
@kpulse360.bsky.social[Union européenne] Deepfakes : le dispositif de transparence de l’AI Act
La Commission européenne impose un marquage obligatoire pour les deepfakes via l’article 50 de l’AI Act, tandi
https://infos.kanibal-studio.com/deepfakes-le-dispositif-de-transparence-de-l-ai-act-7574c4
@nisdra.bsky.socialBilan de la semaine : deux horloges, une seule sonne aujourd'hui. Transparence des systèmes d'IA, en vigueur depuis le 2 août 2026. Haut risque, repoussé au 2 décembre 2027. La première ne demande pas un projet : elle demande une phrase au bon endroit. #AIAct
@radiohack.bsky.social📣 New Podcast! "Agenti AI di OpenAI fuori controllo: escalation di incidenti, indagini e rischi per la sicurezza" on @Spreaker #agentiai #agentiautonomi #ai #aiact #cybercrime #cybersecurity #digitalsecurity #eticaai #incidentiai #innovazione #intelligenzaartificiale #newsai #openai #privacy
@raveneyeglobal.comSince August 2, the EU AI Act requires chatbots and other interactive AI systems to tell users they are not a person. It's a simple transparency rule. Check the welcome message on yours.
@nisdra.bsky.socialFournisseur ou déployeur : la distinction décide de ce que vous devez faire. La PME qui achète un outil d'IA est presque toujours déployeur — et c'est au déployeur qu'il revient d'indiquer qu'un contenu ressemblant à une personne réelle a été généré. #AIAct
@ainewsbulletin.bsky.socialEU AI Act clarifies definition for general-purpose AI models. Rules apply to all providers in the EU market
Sponsor this news channel today. https://www.ProNetGrowth.com/?utm_source=bluesky&utm_medium=post&utm_campaign=news-ai&utm_content=ai
@forsee-research.bsky.social
@liberties.euDid you miss Liberties' recap?
🗳️ Germany’s election fallout
🛰️ A new EU #CivicSpace Tracker
🤖 #AIAct loopholes & the Data Omnibus
⚖️ Our members' updates on #justice, #surveillance & #digitalrights
Catch up here
👉 link.liberties.eu/cca5fe
Subscribe
➡️ link.liberties.eu/52d009
@chochola.czOpenAI did not formally report the RubyGems security incident from May. The EU AI Office has now confirmed it.
For me it's the first documented breach of the AI Act reporting obligation. Advice for companies: documentation isn't enough, you need an incident reporting process.
