Building a Brighter Future: Humans and Artificial Super Intelligence in Transformative Alliances

Karel Capek introduced the world "robot" in 1920 and depicted machines to perform human tasks. The robots in "R.U.R." developed human-like qualities and intelligence. A decade ago, "Superintelligence: Paths, Dangers, Strategies" by Nick Bostrom was influential in questions raising concerns about existential risk from AI. And more recently in 2017, "Life 3.0: Being Human in the Age of Artificial Intelligence" by MIT professor Max Tegmark explored other profound implications of artificial general intelligence (AGI).

For now, leading professionals and companies have embraced AI. Organizations have identified use cases and are enabling the end users to share prompts (For example, the Microsoft Prompt Buddy App). Technology companies are improving #LLMs and foundations with #RAG and exploring other technical improvements as well as running on the edge. New benchmarks are required, yet hundreds of models have been deployed and the most relevant are being tracked. New AI Unicorns will emerge.

Today, most industries have many inefficiencies. According to IDC, beyond 2024, over 40% of core IT spending will shift to AI-related initiatives. I have shared multiple posts on how AI increases the productivity in X. Gartner expects 30% of workers to leverage digital "charisma filters" to improve productivity and career. Workers will commonly use Generative AI: Forrester said that 60% of workers will use their own AI to perform tasks. 44% of business leaders, according to a PWC survey, plan to implement data modernization efforts in 2024 to support GenAI. Beyond companies, AI has plenty use cases for improving our governments and society. Brad Smith summarized many of the more impactful examples in "AI for good". AI with automated decision-making has already rolled out. On the negative side, McAfee forecasts that AI will help cybercriminals manipulate social media and shape public opinion. The EU region is leading AI regulation, I'll write more about that topic in another article.

All eyes are on AGI. Ethan Mollick from Wharton provides a great introduction and explainer of four possible AGI scenarios in this short video. For a very detailed view of the present and future of AGI, Situational awareness is a must read. Before moving along, a reminder for everyone to push ChatGPT beyond simple queries and well-known techniques; consider explicit challenges in your prompts such as "What is the most expensive way to solve this problem?" or "how it can be solved without any technological or monetary constraints". Now that we have all the context on the different phases of AI, lets image the future. I will not focus on the dystopic scenarios, with a high p(doom), but instead in more favorable outcomes.

The term Artificial Super Intelligence (ASI) is being used to describe superior to human intelligence in every field. These ideas for a post-AGI era are my own. I envision a transition from "copilots" to being able to work with multiple of those ASIs or "genius AGIs". Imagine that you could work with Leonardo Da Vinci, Galileo Galilei, Albert Einstein, Nikola Tesla all at the same time, smarter digital versions of them combined with context of other innovators. Imagine that you can compose music with the help and feedback from Johann Sebastian Bach, Ludwig Van Beethoven, and your favorite more recent musicians. taking full advantage of multimodal capabilities. Imagine chatbots of Elon Musk, Jeff Bezos, Bill Gates, Steve Jobs in a panel against your ideas. Imagine not only the first multibillion-dollar one-man organization, but the government running a virtual think tank that can accelerate social, economic, and scientific issues. Likely ASIs will better understand the behavior of whales, birds and other living beings in ways we have not seen, even our own. How will prosthetics, bionics, neural implants or other human enhancements will allow us to interact with AGIs? The singularity is nearer. Not a single optimization algorithm can perform best on all potential problems... Will multiple AGIs compete or collaborate with one another? Does that an even larger threat for humanity or a potential mitigation, since ASIs might be more interested in evolving, "surviving" or becoming symbionts? There will always be physics limitations but how a world with abundant super-intelligence looks like?