(OpenAI reported 1/15 teaspoon in a blog post for a typical prompt for ChatGPT) Here is the report:
@emollick
-
Real World AGI Signals vs Social Media Hype Discrepancy
By
–
In sum, the real world signals of how seriously people are taking AGI differs from the impression you get of the possibility of AGI on X. Whether that means that people close to AI development know more or that they are vastly overestimating AGI possibilities is the key question
-
AGI Integration Challenges: Organizational Timeline and Adoption Barriers
By
–
The “it won’t matter much” argument is that AGI will still require huge amount of time and effort to integrate into organizations and workflows, no matter how smart, resulting in years of work & gradual changes. Needless to say, it also suggests no true AI superintliigence soon.
-
AGI Development: Expected Changes in Investment and Corporate Strategy
By
–
AGI being “a machine better than average humans at every intellectual task” or related concept. You would expect to see changes in investment patterns & corporate strategy, crash R&D efforts to set up organizations to benefit from early AGI, policy structures put into place.
-
Industry Leaders Show Limited Confidence in Near-Term AGI Timeline
By
–
For what it is worth, few industry leaders, less than a half-dozen companies & no policy-making bodies are taking actions that suggest that they expect AGI is really a few years away. This may be because they don’t believe it or they think it won’t matter much in the medium term
-
ChatGPT water consumption corrected by OpenAI data
By
–
This chart is apparently already out of date and overemphasizes water use, according to the numbers OpenAI released, 300 average ChatGPT queries is equivalent to 20 tablespoons of water, not 1 gallon. https://
x.com/mattyglesias/s
/mattyglesias/status/1947611012359762255
… -
AI strengths in data compilation and analysis tasks
By
–
I often have a few projects going at a time, but I only use it for areas where I understand the task & the output I want. In general, it is very good at data compilation and analysis, its mistakes feel more human-like than traditional hallucination, confusion over columns, etc.
-
ChatGPT Agents as Interns: Practical AI Oversight and Efficiency
By
–
I am finding ChatGPT agents to be useful. They are a better fit with the "intern" analogy than any former AI – requiring oversight, still saving lots of time overall. For example, I update an AI cost/performance chart frequently. The agent did all the grunt work, with guidance.
-
LLM Architecture: RL Thinking Systems Debate
By
–
I don't think that is correct. It is an LLM with an RL "thinking" system on top. I would not call that a higher level system, but we may just be splitting hairs on definitions.
-
LLMs Solve Hard Math Problems Through Generalization
By
–
It wasn't just OpenAI. Google also used a general purpose model to solve the very hard math problems of the International Math Olympiad in plain language. Last year they used specialized tool use Increasing evidence of the ability of LLMs to generalize to novel problem solving