- - RT @gerdschenkel: DALL·E 2 AI image-generator is free for all: now I can make photos without a camera! https://t.co/roDLW1OeDD dlvr.it
- - RT @mpesce: AI Hyperopia by Matt Calder https://t.co/oWdM8A53Uu An excellent analysis of what's right - and often, quite wrong - about th… link.medium.com
- - AI Ethics And AI Laws Reveal Troubling Concerns From Tesla’s AI Day Showcase And The Ever Expanding AI Ambitions Of Elon Musk https://t.co/l0QiQLjqp5 forbes.com
- - RT @PatrickCMiller: NIST to establish online center for trustworthy and responsible AI resources https://t.co/8pVDX3S1H2 bit.ly
- - The Yield: How to feed the world without ‘wrecking the planet’ https://t.co/SIQev6abqy news.microsoft.com
- - AI Stock Images: The Uncanny Valley from ShutterStock to StockAI https://t.co/a6lcm3NSRS prototypr.io
- - DALL-E 2 AI image generation is coming to Microsoft Bing https://t.co/SG3MlQ2cLe windowscentral.com
- - RT @technollama: Well, this is big. Microsoft adds DALL-E capability to Office. https://t.co/JEh5cDQ8y8 theverge.com
- - Behind 'miracle' AI is an army of 'ghost workers' — and they're speaking out about Appen https://t.co/6HLL7Gp9uu abc.net.au
- - RT @ThisIsDK999: I just published It’s the Little Things : Breaking an AI https://t.co/RxNatSC0MJ #bugbounty #hacking #infosec infosecwriteups.com
- - Prompt Injection: A Critical Vulnerability in the GPT-3 Transformer, and How We Can Begin to Solve It https://t.co/fysdZA5PVT preamble.com
- - ImpNet: Imperceptible and blackbox-undetectable backdoors in compiled neural networks https://t.co/WdVG9IoNud arxiv.org
- - RT @_akhaliq: Imagic: Text-Based Real Image Editing with Diffusion Models abs: https://t.co/xRW6F6w2ZG https://t.co/wGifY74i4w arxiv.org
- - Avatar AI - Create your own AI-generated avatars https://t.co/xwxgKuq5hl avatarai.me
October 2022
October 2022 is defined by DALL-E's rapid democratization — going free, integrating with Microsoft Bing, and embedding into Office 365 — alongside growing awareness of the hidden human labor behind 'miracle AI', with Appen workers speaking out about exploitative data labelling conditions. Security research deepens around prompt injection vulnerabilities and imperceptible backdoors in neural networks, while NIST moves to establish a trustworthy AI resource center, reflecting early regulatory intent. The parallel tracks of capability expansion and accountability debate are in full tension, with AI stock imagery challenging visual authenticity standards and Imagic-style diffusion editing opening new creative possibilities under increasingly contested ethical ground.
We're not responsible for the content of these links.