Pourquoi je déconseille formellement de faire des études de médecine en 2025 1 Les études sont beaucoup trop longues : 10 ans pour être généraliste 2 La formation est TOTALEMENT inadaptée à l’ère de l’IA 3 ChatGPT est déjà 4 fois meilleur que les médecins pour faire un
GENERATIVE AI
-
Kimi K2 Model Exhibits Significant Hallucination Issues
By
–
Kimi K2 really hallucinates a lot in my limited testing so far, and is very happy to make up new details if it judges that it would improve the punch of a paragraph.
-
Alibaba Qwen3 Models Released: Thinking and Translation Capabilities
By
–
Alibaba's Qwen team dropped two more models: —A Qwen3-Thinking update, now competitive with Gemini 2.5 Pro, o4-mini, and DeepSeek R1 across multiple benchmarks
—Qwen3-MT, an AI translation model with support for 92+ languages -
Runway Releases Aleph for AI-Powered Video Post-Production Editing
By
–
Runway dropped Aleph, a new model for AI-powered post-production editing
— The Rundown AI (@TheRundownAI) 28 juillet 2025
It can perform a range of edits on video inputs, including adding, removing, and transforming objects, producing new angles of a scene, modifying style and lighting, and morepic.twitter.com/1SRiqRkmHwRunway dropped Aleph, a new model for AI-powered post-production editing It can perform a range of edits on video inputs, including adding, removing, and transforming objects, producing new angles of a scene, modifying style and lighting, and more
-
Tencent Releases Hunyuan3D World Model for 3D Generation
By
–
Tencent released Hunyuan3D World Model 1.0, a new open-source world generation model
— The Rundown AI (@TheRundownAI) 28 juillet 2025
It creates interactive, explorable 3D worlds from image or text prompts and is also compatible with CG pipelines for full editability & simulationpic.twitter.com/Px5SIyoDDGTencent released Hunyuan3D World Model 1.0, a new open-source world generation model It creates interactive, explorable 3D worlds from image or text prompts and is also compatible with CG pipelines for full editability & simulation
-
Meta Hires OpenAI Researcher Shengjia Zhao as Chief Scientist
By
–
Meta Superintelligence Labs finally found its chief scientist: ex-OpenAI researcher Shengjia Zhao Zhao helped create OpenAI's GPT-4, o1, o3, 4.1, and mini models, and will set MSL’s research direction alongside chief AI officer Alexandr Wang
-
Model 4.5 Excels at Rewriting While Preserving Original Style
By
–
4.5 is the only model I trust with writing, esp re-writing without changing the style – all others don't understand the task
-
Sam Altman: AI will enable software creation using plain language
By
–
Sam Altman affirme que l’IA permettra bientôt à n’importe qui de créer un logiciel simplement en le décrivant en anglais courant.
— VISION IA (@vision_ia) 28 juillet 2025
Toute personne ayant une idée pourra la formuler, et l’IA s’occupera de la construire, de la déboguer et de la déployer à sa place.
On a commencé à… pic.twitter.com/Azw6t9LFmbSam Altman claims that AI will soon allow anyone to create software simply by describing it in plain English. Anyone with an idea will be able to articulate it, and the AI will take care of building, debugging and deploying it on their behalf. We have started to
-
When Will AI Model Sizes Reach Avogadro Scale?
By
–
How long before model sizes are measured in Avogadros (multiples of 6 x 10^23)?
-
25 AI Customer Success Stories: ROI and Value Creation Examples
By
–
25 customer success stories of #AI's impact on value creation, cost reduction, and revenue generation: https://
cloud.google.com/transform/25-o
f-my-favorite-roi-customer-stories-gen-ai/
… by @googlecloud