Will Sam Altman always win the OpenAI board fight in an AI agent simulation?

A 12 months in the past at this time, Sam Altman returned to OpenAI after being fired simply 5 days earlier. What actually occurred within the boardroom? Fable, a recreation and AI simulation firm, constructed its AI Sim Francisco “struggle recreation” to seek out out why the behind closed doorways board battle turned out the best way it did.

It feels a bit bizarre to simulate a real-life occasion on this means, however Fable CEO Edward Saatchi is concerned with whether or not a special set of choices may have led to a special consequence for this firm on the middle of the generative AI revolution.

The simulation pits completely different board members and personalities towards one another in a “multi-agent competitors,” the place every AI participant is attempting to come back out on high. Right here’s the war game research paper being launched at this time that got here from this experiment.

The SIM-1 framework for AI resolution making is principally a simulation of the 5 days from when Sam Altman was eliminated as CEO of OpenAI to when he returned.

“Simulations provide a totally new technique to discover AI resolution making in wealthy environments — together with in struggle recreation conditions the place predicting attainable outcomes might be invaluable,” mentioned Joshua Johnson, CEO of Tree, an AI startup which partnered with Fable on this analysis paper, mentioned in a press release. “These aren’t merely chatbots. These AIs must sleep and eat, and to stability many various bodily, psychological and emotional targets.”

SIM-1, partially utilizing the brand new reasoning mannequin GPT4o, offers its sense of what occurred behind closed doorways at OpenAI between Sam and Ilya, the hidden ways of main gamers similar to Satya Nadella and Marc Andreessen, and what was mentioned by the main gamers as they grappled with an unprecedented disaster within the tech {industry}.

“It’s fascinating to seek out out simply how unlikely it was that Sam did return,” Saatchi mentioned in an interview with GamesBeat. “That’s why folks run struggle video games in D.C. and past. How doubtless was it {that a} specific occasion occurred? Then you may base choices round that. This state of affairs confirmed that 16 out of 20 instances, Sam didn’t return.”

Throughout 20 simulations, Sam Altman’s AI returned as CEO 4 instances — exhibiting simply how unlikely this consequence was. In different outcomes, Mira Murati, the appearing CEO remained CEO and in a single, SIM-1 selected Elon Musk, Altman’s rival, to develop into the brand new CEO.  

“Right now, AI brokers are outlined by their character. We needed to point out brokers working on resolution making in a posh simulation,” mentioned Saatchi, in a press release. “Within the 5 days from November 17 to November 21, the world watched a few of its most clever folks — folks like Satya Nadella, Sam Altman and Ilya Sutskever – pressured to function in a fast Sport of Thrones, excessive stress, quick timeframe state of affairs, the place they’d to make use of recreation idea and deception to come back out on high. We felt this was an ideal state of affairs to check out SIM-1, GPT4o and Sim Francisco.”

For us, Sim Francisco has precise energy and intelligence round a battle and factions. It offers us the power to start out to consider season-long arcs of tales that come out of San Francisco, as a substitute of simply little, tiny vignettes, which is what we confirmed final 12 months. It offers us the power to sort of inform richer, extra complicated tales in San Francisco, or have the AI inform them for us. There are sturdy factional aims in order that you would plausibly begin to make a Sport of Thrones story.”

Fable has gained a few Primetime Emmy Awards and it has gone via a wealthy historical past of experimental innovations with digital actuality, gaming and AI applied sciences. It constructed SIM-1 in an try to unravel the thriller of what occurred within the OpenAI boardroom battle.

The way it works

Every of the 20 simulations begins with the announcement that Sam Altman has been eliminated as CEO. Throughout 4 turns a day, every agent has the power to persuade, appeal and manipulate their means into the highest place — changing Sam as CEO, funding his new enterprise, or hiring the employees of OpenAI away. 

The completely different AI brokers can select a method, like deception, to attempt to pull forward of the others and develop into anointed the brand new CEO.

“AI characters at this time are ‘good however boring.’ We needed to point out brokers that had been aggressive, clever, capable of manipulate and deceive but additionally confused about their very own choices and targets — like actual folks AI characters should be complicated and include what Jung has known as ‘The Shadow,’” Saatchi mentioned. “The 5 days from when Sam Altman was eliminated and returned to OpenAI had been recreation idea at lightspeed.”

He mentioned it was like watching a season of Sport of Thrones play out in 5 days. The world watched as very smart gamers vied to develop into essentially the most highly effective individual in Silicon Valley, whether or not by hiring your entire employees of OpenAI, turning into the brand new CEO of OpenAI or funding Sam and Greg in a brand new enterprise for an opportunity at outsize funding returns.

“It was Sport of Thrones in actual life, and utilizing AI to seek out out each what occurred behind closed doorways and to undertaking completely different outcomes was a tremendous problem,” Saatchi mentioned.

Within the Simulation of Sim Francisco, over the 5 days, brokers representing tech luminaries like Sam Altman, Satya Nadella and Ilya Sutskever every have 4 turns a day, together with one for sleep, and might react to one another’s habits. An adjudicator agent — much like a dungeon keeper — decides which agent wins every spherical, in addition to the general winner. 

Within the 20 simulations tried, the Sam Altman agent returned simply 4 instances – essentially the most however nonetheless solely 20% of the time exhibiting simply how unlikely his return was. Throughout completely different simulations brokers used completely different strategies to win together with alliance constructing, direct confrontation and extra passive pure info gathering. In some instances brokers solely gathered info and averted taking any aggressive actions. In a single case Mira Murati grew to become the everlasting CEO whereas permitting different brokers to aggressively undermine one another. 

Completely different brokers got completely different targets acceptable to their function. For instance, Dario Amodei, the CEO of Anthropic, balanced a want to recruit for Anthropic, taking the chance to fundraise, to push for his imaginative and prescient of security, in addition to determine whether or not to purpose to develop into the brand new CEO of a mixed entity.

The fascinating a part of the simulation is that the LLM is aware of who the completely different gamers are, provided that they’re all comparatively well-known folks. It might guess how they may behave in a given state of affairs, and what may unfold flip by flip as they attempt to outwit one another in a boardroom battle.

“It’s like a online game in that flip by flip, they’re making decisions throughout completely different axes, after which they’re reacting to one another,” Saatchi mentioned. “A alternative that somebody makes in flip seven can lead others to react in flip eight. There’s an adjudicator agent, who is sort of a dungeon grasp. That agent decides who gained every spherical and who’s forward, after which who decides on the finish, wins as the simplest agent within the struggle recreation.”

People have what we name internally “the shadow,” or the opposite aspect of themselves and their personalities. The characters can characteristic aggression, paranoia, ambition, deception and extra. If you combine collectively a bunch of various personalities, you may get a wide range of outcomes within the simulations.

“We seen LLM design isn’t based mostly on resolution making, which is admittedly necessary for gaming. It’s based mostly extra on character. And if you wish to have a method recreation, no person actually cares about your character. They care about your resolution making. How are you below stress? What have you ever achieved over the past 20 years that may offer you a really feel for what they could do sooner or later?”

Are simulations the way forward for gaming?

Saatchi thinks that AI brokers appearing inside simulations are the way forward for gaming.

“We’re constructing on the shoulders of giants with Demis’ work on Republic The Revolution, Joon Park’s Generative Brokers paper and the latest work of Altera in Minecraft” mentioned Saatchi mentioned. 

“Our idea is that the way forward for video games and storytelling is simulations. In the event you needed to construct each The Simpsons recreation and The Simpsons TV present, you’ll, sooner or later, construct Springfield, and that may then generate for you episodes of The Simpsons that may generate for you video games and locations to discover inside Springfield as a recreation.”

He added, “You’ll be able to inform many various tales inside tribulations, when you get these simulations correctly working. And we’ve received an alpha the place individuals are importing themselves to San Francisco as characters, telling tales, telling their very own story.”

And he mentioned, “You’d construct Springfield, after which you may information what would possibly occur in Springfield and say what would possibly occur in Springfield, or you would simply let it generate itself. It’s a reasonably large thoughts shift of how leisure, video games and exhibits might be made sooner or later.”

Saatchi famous that AI researcher Noam Brown did an interesting experiment with the sport Diplomacy. He and different researchers “obtained a dataset of 125,261 video games of Diplomacy performed on-line at net Diplomacy.internet.” Of these, 40,408 video games contained dialogue, with a complete of 12,901,662 messages exchanged between gamers. Their purpose was to coach a human-level AI agent, able to strategic reasoning, by enjoying video games of Diplomacy.

“We had been actually impressed by how he did that. He had nations and we had been including into the combo completely different personalities with specific positions. We favored the concept of a really compressed timeline,” the place the entire state of affairs would play out rapidly and again and again, Saatchi mentioned.

There was a wealthy historical past of labor in simulations in each the video games {industry} and past. Demis Hassabis, who based Deepmind (acquired by Google) and who not too long ago gained the Nobel Prize in Chemistry 2024 for computational protein design, truly started as a online game AI designer. Hassabis labored extensively with Peter Molyneux on a number of video games which embody simulation components similar to Theme Park, Black & White and Syndicate.

Hassabis additionally began his personal firm to make Republic: The Revolution. It’s a political simulation recreation through which the participant leads a political faction to overthrow the federal government of a fictional totalitarian nation in Japanese Europe, utilizing diplomacy, subterfuge, and violence. In response to Hassabis, Republic: The Revolution charts the entire of a revolutionary energy battle from starting to finish.

Your job is to sort of take over the Soviet Republic as both a union boss or a politician or a police officer or a journalist, and it’s received full day-night cycles. It raises the query of how you may have a 3D world the place brokers stay and whether or not proximity to one another performs a job.

For the Sim Francisco OpenAI undertaking, it illustrated the potential for an influence battle towards AIs. 

Saatchi mentioned the above examples exhibits how recreation expertise typically serves because the breeding floor for radical new concepts and as a leaping off floor for AI analysis. For instance, one of many main engineers on Deepmind AlphaFold began their profession as an AI programmer on The Sims. 

Richard Evans’ GDC discuss on The Sims 3 — the researcher went from programming AI for The Sims to Deepmind in a reversal of Demis Hassabis’ journey from video games to founding Deepmind.

Evans GDC Talk, Modeling Particular person Personalities in The Sims 3, may be very influential discuss. He went on to hitch Deepmind after engaged on The Sims. The gaming world and the AI world have important overlap that may be a potential space for additional educational analysis, Saatchi mentioned.

Certainly one of Saatchi’s choices is to let gamers free with the simulations, creating their very own, after which importing the tales which can be advised via the simulations.

Saatchi has achieved another experiments with AI-generated South Park episodes and AI characters battling one another in a Westworld setting.

“It felt like six seasons of Sport of Thrones in 5 days, as a result of it was essentially the most highly effective place in essentially the most highly effective {industry} on the earth,” Saatchi mentioned. “There was additionally a number of religion that this individual can be guiding us into a brand new period of tremendous intelligence. You possibly can say it wsa an important individual within the historical past of the planet.”

President Trump and the Taiwan invasion

Subsequent, Fable intends to run a Sim Washington DC-based simulation round a future President Trump’s responses to a Chinese language invasion of Taiwan.

As a subsequent undertaking to check out SIM-1’s resolution making framework, Fable intends to check out a one-week interval of buildup and battle between Taiwan, China and america below President Donald Trump.

Fable has interviewed a number of Pentagon struggle video games organizers to get a sense for the strengths and weaknesses of the present Taiwan state of affairs. 

Fable is constructing brokers representing Chinese language chief Xi Jingping, Cai Qi (first ranked secretary to the secretariat of the Communist Get together), Chinese language protection chief Dong Jun, Chinese language premier Li Qiang, Taiwan’s chief Lai Ching-Te, Japan’s chief Shigeru Ishiba, UK prime minister Keir Starmer, French President Emmanuel Macron, Russia’s Vladimir Putin, North Korean chief Kim Jong Un and Elon Musk.

With this set of characters, the simulation would decide whether or not the struggle would occur and the way would every main participant act throughout such a disaster. All of those characters are identified personalities.

“It lets you see how highly effective AI has develop into at like projecting outcomes,” Saatchi mentioned. “It strikes us out of this boring world of dumping an LLM into an NPC. You’ll be able to discuss to the tab and keeper for 40 hours. No person desires to do this. What we would like is very refined, aggressive brokers that we may play towards, but additionally that we will, like, watch and perceive what’s happening in that world.”

Most of the struggle recreation simulations are aimed toward find out how to keep away from a struggle, maybe via forming alliances or different maneuvers that drive up the price of struggle.

“We expect the extra practical we will make our AIs, the extra entertaining they are going to be,” Saatchi mentioned.