The Guardian view on DeepSeek and a global AI race: geopolitics, innovation and the rise of chaos https://
theguardian.com/commentisfree/
2025/jan/26/the-guardian-view-on-a-global-ai-race-geopolitics-innovation-and-the-rise-of-chaos?utm_source=chatgpt.com
…
GENERATIVE AI
-
DeepSeek and global AI race: geopolitics, innovation, chaos
By
–
-
AI Economics Shift: Inference and RL Drive Future Costs
By
–
And for those arguing this means less reliance on compute—it actually means more.
 The game is shifting. The biggest costs aren’t just in pre-training on human data anymore.
 It’s on inference and new training techniques like reinforcement learning to drive the next -
DeepSeek Export Bans Real Story Beyond Cheating Accusations
By
–
🇺🇸🇨🇳 The uproar over #DeepSeek “cheating” is missing the real story.
— Nina Schick (@NinaDSchick) 4 février 2025
Yes, even if they had access to 50,000 NVIDIA H100s—$1.5–2B worth of compute—despite export bans.
But let’s be real: U.S. export bans don’t stop compute from flowing.
20% of NVIDIA’s revenue comes from… pic.twitter.com/jSp0hGiYPbThe uproar over #DeepSeek “cheating” is missing the real story. Yes, even if they had access to 50,000 NVIDIA H100s—$1.5–2B worth of compute—despite export bans. But let’s be real: U.S. export bans don’t stop compute from flowing. 20% of NVIDIA’s revenue comes from
-
AI-Augmented Journalism: Accountability, Impact, and Risks
By
–
How is AI transforming journalism? From uncovering critical accountability stories to reporting on AI’s impact on vulnerable communities, this workshop will discuss the need, opportunities, and risks for AI-augmented journalism. Live on Feb. 13: https://
bit.ly/40WX0dr -
AmbiStack System Uses AI and Reinforcement Learning for Package Placement
By
–
Great summary Lukas! The new AmbiStack system uses @AmbiRobotics’ PRIME-1 Generative AI model trained on 20,000 hours of real robot data for perception and a Reinforcement Learning model for highly efficient package placing (a 3D variant of playing Tetris;). https://t.co/DxPmvXuQuM
— Ken Goldberg (@Ken_Goldberg) 4 février 2025Great summary Lukas! The new AmbiStack system uses @AmbiRobotics
’ PRIME-1 Generative AI model trained on 20,000 hours of real robot data for perception and a Reinforcement Learning model for highly efficient package placing (a 3D variant of playing Tetris;). -
When Will AI Surpass the Smartest Human Intelligence?
By
–
Vus les progrès foudroyants de l’IA depuis début décembre 2024 Quand pensez vous que l’IA dépassera l’homme le plus intelligent sur terre ? 2025 comme l’affirme @elonmusk 2026 comme @StephaneMallard 2027 comme Ray Kurzweil
2028 comme pense @sama -
Trust Emerges as Key Factor Beyond Foundational AI Models
By
–
"In this early rush for gold in the AI market, we’ve become so fixated on the foundational models that we’ve forgotten that the real, biggest market has yet to emerge, and trust will be central to winning it." My weekly column for @InfoWorld
-
Synthetic Data and AI Protect Right Whales from Boat Strikes
By
–
What does #SyntheticData have to do with balancing commerce and conservation? See how SAS combined this #GenerativeAI tech with Machine Learning to help #DataForGood partner Fathom Science validate its efforts to protect right whales from boat strikes.
-
Gemini 2.0 Pro now in Google Workspace
By
–
BREAKING : Gemini for Google Workspace support page got updated mentioning Gemini 2.0 Pro "Gemini 2.0 pro is our most capable AI model widely available today – Far more capable at reasoning, following instructions, coding, and creative collaboration."
-
Beyond Generative AI Hype: Culture and Data Game-Changers
By
–
Take a listen to my latest #audio #podcast "Beyond The Generative AI Hype: Why Culture And Data Are The True Game-Changers In 2025". If you want to move beyond the AI hype and understand what truly drives success with generative AI, this conversation will give you the