AI-augmented predictions, mapping intelligence, AI in soccer, Mormon AI, and more

March 23, 2024

“Some people call this artificial intelligence, but the reality is this technology will enhance us. So instead of artificial intelligence, I think we'll augment our intelligence.” – Ginni Rometty

The biggest of the many problems in our discussions of “artificial intelligence” and more generally intelligence, is that almost no-one can give a good definition of what they mean when they say “intelligence”. Our reference point has been what people can do, some better than others. Now we have to pull back to understand intelligence from a broader frame.

Below I have shared my Mapping Intelligence framework, which aims to help people think about intelligence by offering fundamental dimensions and prominent definitions.

The Humans + AI Explorers Community has gained over 40 members in the last week, many of them world leaders in the topics we discuss. Join us, it’s free!

Ross

 

📖In this issue

  • Mapping Intelligence framework

  • AI is an adept soccer tactician

  • AI is more persuasive at personalized persuasion than humans

  • GPT-5 to launch mid-year with strong expectations

  • Video AI nurses are available at $9 per hour

  • Philipp Schoenegger on AI-augmented predictions, improving human decisions, LLM wisdom of crowds, and how to be a superforecaster

  • Leaders and employees have very different outlooks on the impact of GenAI on work

 

 

🖼️ Framework

Mapping intelligence

In all the discussion of intelligence, human and artificial, people are rarely stopping to clarify what they mean by “intelligence”. This framework maps the six major dimensions of intelligence as usually defined, along with an array of prominent definitions illustrating these themes.

 

👩🤖Humans + AI update

Google DeepMind releases AI football tactician in partnership with Liverpool FC

“We introduce TacticAI: an artificial intelligence (AI) system that can provide experts with tactical insights, particularly on corner kicks, through predictive and generative AI. Despite the limited availability of gold-standard data on corner kicks, TacticAI achieves state-of-the-art results by using a geometric deep learning approach to help create more generalizable models.”

LLMs that personalize their arguments can persuade better than humans

LLMs are around as good as humans at persuading people, but if given only basic demographic data they are substantially better at convincing people to change their minds.

Mormon Church’s guidelines for AI

“‘Light and truth’ are a scriptural definition of intelligence,” Elder Gong said, speaking of Doctrine and Covenants 93:36. “While generative artificial intelligence may be quick to offer information, it can never replace revelation or generate truth. If something does not feel right or is inconsistent with what you know is true, seek to discern before believing.”

 

🔥Hot news in AI

  • Japanese startup Sakana AI uses “model merging”, using multiple GenAI models to evolve better systems - Reuters

  • GPT-5 said to be launching in mid-year alongside AI agents; corporate clients have received demos - Business Insider

  • At its annual conference Nvidia announces the next generation Blackwell AI chips and GR00T humanoid robot model - CNBC

  • Microsoft hires DeepMind and Inflection AI co-founder Mustafa Suleyman to run Microsoft AI - Microsoft

  • Nvidia and Hippocratic AI offer GenAI nurses for $9 an hour to offer medical advice in video calls (see video)

 

🎙️Latest podcast episode

Philipp Schoenegger on AI-augmented predictions, improving human decisions, LLM wisdom of crowds, and how to be a superforecaster

What you will learn

  • Exploring the intersection of AI and human decision-making 

  •  The catalytic effect of ChatGPT on modern research 

  •  The fundamentals of AI-augmented forecasting 

  •  Unpacking the wisdom of AI crowds 

  •  The journey to becoming a superforecaster 

  •  Navigating the blend of human intuition and AI computation 

  •  Insights into the future of AI-enhanced judgment 

 

💡Resources and insights

Leaders and employees have very different outlooks on the impact of GenAI on work

“Despite 94% of people saying they are ready to learn new skills to work with gen AI, only 5% of organizations are actively reskilling their workforce at scale.” An interesting report from Accenture “Work, Workforce, Workers Age of Generative AI” explores current state and future based on an extensive survey.