Sam Altman departs Open AI https://
x.com/evacide/status
/evacide/status/1725617532412068025
…
@deeplearn007
-
Sam Altman Departs OpenAI Leadership
By
–
-
Scaling AI to Reshape Work Requires Memory and Reasoning
By
–
Eventually but only when it can be scaled in a cost and resource efficient way including energy consumption costs. To really reshape work it will require memory and causal reasoning too. @MorganStanley @BetaMoroney
-
GPT 4.5 and 5 launch hype cycle predictions for 2030s
By
–
Revisit this discussion in the 2030s albeit the social media hype will probably return when GPT4.5 and then 5 are launched.
-
Energy Efficiency and Carbon Footprint in AI Model Development
By
–
We need to evaluate progress in AI models also in terms of energy consumption, costs of scaling and carbon footprint. More efficient solutions will be key.
-
Sustainable and Scalable Generative AI Drives Business Value
By
–
The true value from Generative AI for business will arrive we when we make the AI models more sustainable and scaleable across the economy.
-
Beyond Transformers: Bio-Inspired AI for Future Innovation
By
–
There were many concepts and dreams a long time ago but until Transformers & GANs took off much of the AI tasks including Generative AI did not work. The question is where do we go next? I believe the innovation should be towards biologically inspired more resource and energy
-
Ethics in Theory and Practice: Regional and Organizational Perspectives
By
–
Ethics in theory is simple but in reality it is more complex. We have to start by asking why we ethics? This will vary by region / society and from whose perspective? However, a belief in doing the right thing is better for the organisation and the individual in the long-run.
-
Biologically Inspired Approaches for Advanced Logical Reasoning AI
By
–
Logical reasoning is important and a missing part to attain truly advanced AI. I am not sure current Neurosymbolic approaches will get us there. I am more optimistic about biologically inspired approaches that look to the role of our own brains for the solution.
-
POTUS AI Policy: Data Access, Bias, and Energy-Efficient Computing
By
–
IMO it’s great to see @potus take a proactive stance on good foundations for advanced AI. We also need action on data access, bias, incentives for AI that consumes less energy (lower carbon footprint) with lower computational resources to scale around the edge (devices). This
-
Transformer AI: Multimodal Progress and Energy Cost Challenges
By
–
Generative AI in particular applying Transformer architecture is a step forward by bringing multimodal, multitasking AI – moving beyond narrow AI but not human level AGI. However, it is not a long-term solution as the computational costs are very high resulting in high energy