The Possibilities of AI [Entire Talk] - Sam Altman (OpenAI)

In a conversation at Stanford University's ETL seminar, Sam Altman, CEO of OpenAI, discusses his journey from a Stanford undergrad to leading OpenAI, emphasizing the importance of resilience and adaptability in the rapidly evolving AI landscape. He reflects on his consistent traits of excitement, optimism, and curiosity, and underscores the unparalleled opportunities for today's entrepreneurs in AI research and development. Altman candidly shares his perspective on the future challenges of AI, including responsible deployment and the potential societal impacts. He advocates for iterative product releases to allow society to adapt alongside AI advancements and stresses the significance of building larger, more capable AI systems, while also acknowledging the unknowns and the necessity for ethical considerations in AI's progression.

• Sam Altman is the co-founder and CEO of OpenAI, the research and deployment company behind ChatGPT, DALL-E, and Codex.
• OpenAI was initially founded as a non-profit research lab with the mission to build general-purpose artificial intelligence (AI) that benefits all humanity.
• OpenAI's ChatGPT set a record for the fastest-growing app in history, reaching 100 million active users just two months after launch.
• Sam Altman has been recognized as one of Time's 100 most influential people in the world and CEO of the year in 2023.

"OpenAI is the research and deployment company behind ChatGPT, DALL-E, and Codex."

This quote highlights the major products developed by OpenAI, emphasizing the company's influence in the AI space.

• Sam grew up in the Midwest, attended Stanford University, and took the ETL seminar as an undergrad.
• He joined Y Combinator with a social mobile app company called Loopt, which was later acquired.
• Sam became the president of Y Combinator from 2014 to 2019 before co-founding OpenAI in 2015.
• He lives with his husband in San Francisco and splits his time between San Francisco and Napa, and is also a vegetarian.

"Sam joined the inaugural class of Y Combinator with a social mobile app company called Loopt, which then went on to raise money from Sequoia and others."

This quote summarizes Sam Altman's early entrepreneurial success and involvement with Y Combinator, a notable startup accelerator.

• When asked to describe his younger self at Stanford with three words, Sam chose "excited, optimistic, and curious."
• Sam believes that the same qualities still apply to him today, indicating a consistent thread in his character despite changes in the world.
• He views the upcoming changes in the next 19 years as significant and sees it as an opportunity for impact through starting companies or conducting AI research.

"Excited, optimistic, and curious."

This quote represents the three words Sam Altman used to describe his younger self, reflecting his mindset as a Stanford undergrad and his consistent approach to life and work.

• Sam considers the current era as an incredibly lucky time to be coming of age, especially for starting companies due to the transformative potential of AI.
• He suggests that students should determine where they want to contribute and pursue it with determination.
• Sam advises against staying a student if one is inclined towards jumping into research or starting a company, emphasizing the unique learning experience from actual startup involvement.

"I would feel incredibly lucky, and I would be determined to make the most of it."

This quote conveys Sam Altman's view on the current opportunities available for young entrepreneurs, encouraging them to seize the moment and make a significant impact.

• Sam would choose to work on meaningful AI research if he were a student again, with a preference for industry due to access to significant computational resources.
• He acknowledges that while academia is a valid path, his personal inclination leans towards industry, particularly in areas where large-scale computing is crucial.

"I think to you kind of need to be the place with so much compute."

Sam Altman's quote indicates the importance of computational power in AI research and his rationale for preferring industry over academia for AI endeavors.

• Sam stresses the importance of charting one's own course and being skeptical of obvious startup ideas that many others are likely to pursue.
• He encourages aspiring entrepreneurs to trust themselves, come up with non-consensus ideas, and learn by doing, rather than seeking advice on what startup to start.

"You kind of just learn how to run a startup by running a startup."

This quote emphasizes the experiential learning process inherent in entrepreneurship and the value of diving into startup creation.

• Sam refrains from specifying which AI challenges are ripe for startups, suggesting that by the time such opportunities are evident, they may not be as promising for new ventures.
• He believes that impactful careers require individuals to come up with their own ideas and pursue non-obvious paths.
• Sam mentions that OpenAI's non-consensus approach during its inception is now seen as the obvious route, illustrating the shift in perception over time.

"I think learning to trust yourself and come up with your own ideas and do the very non-consensus things like when we started OpenAI, that was an extremely non-consensus thing to do."

This quote encourages entrepreneurs to have confidence in their unique ideas and to venture into non-conventional areas, as exemplified by OpenAI's early days.

• Sam is currently focused on how to build really big computers and considers AI infrastructure to be a critical input to the future.
• He believes that addressing the entire ecosystem, including energy, data centers, chips, chip design, and networks, is essential rather than focusing on individual components.
• Sam also contemplates the challenges of integrating advanced AI with PhD-level intelligence into products that positively impact society.

"We're definitely wrestling with how we, when we make not just like grade school or middle schooler level intelligence but like PhD level intelligence and beyond, the best way to put that into a product."

Sam Altman's quote acknowledges the ongoing efforts at OpenAI to not only develop advanced AI but also to consider its integration into society in a beneficial manner.

• Open AI is seen as providing valuable tools for building the future.
• The speaker believes in the ingenuity of people to figure out how to use technology.
• The focus is on iterative deployment, allowing society to co-evolve with technology.
• The company aims to give people and institutions time to adapt and integrate new tools into their lives.
• Open AI's strategy is to ship imperfect products, learn from feedback, and improve.
• The speaker emphasizes that surprises in AI deployment are undesirable; gradual rollout is preferred.

"I kind of think that giving people really capable tools and letting them figure out how they're going to use this to build the future is a super good thing to do and is super valuable."

• The quote highlights the speaker's belief in empowering people with technology to shape the future.

"It's important given what we believe is going to happen to express our view about what we believe is going to happen...and let Society co-evolve with the technology."

• The quote emphasizes the importance of sharing Open AI's vision and allowing society to adapt alongside the technology's development.

"We ship imperfect products...but we have a very tight feedback loop. And we learn and we get better."

• The quote stresses the iterative nature of Open AI's product development, valuing progress through continuous learning and improvement.

• The speaker is not concerned about Open AI's spending or the need for an immediate monetization strategy.
• The belief is that as long as the trajectory is towards creating value for society, the investment is justified.
• The focus is on the long-term potential of artificial general intelligence (AGI) to be worth the expense.

"Whether we burn 500 million a year or 5 billion or 50 billion a year I don't care I genuinely don't as long as we can...stay on a trajectory where eventually we create way more value for society than that."

• This quote reveals the speaker's indifference to the costs associated with developing AGI as long as it leads to significant societal value.

• By 2030, the speaker envisions having a new powerful tool (AGI) but believes life will continue with its usual rhythms.
• The speaker predicts that while AGI will bring changes, many aspects of life will remain unchanged.
• The focus is on the balance between revolutionary technological advancement and the continuity of everyday life.

"It would feel amazing if we got to teleport forward six years today and have this thing that was like smarter than humans in many subjects...and yet...the Sun keeps rising, people keep having their human dramas; life goes on."

• The quote suggests that despite significant technological advancements, fundamental aspects of human life will persist.

• The speaker believes the current definition of AGI is too vague and open to misinterpretation.
• There is an acknowledgment that the timeline for achieving AGI is uncertain.
• The speaker suggests that even with significant advancements in AGI, the impact on global metrics like GDP growth may not be immediately apparent.

"I think we need a more precise definition of AGI for the timing question...it's too loose of a definition, there's too much room for misinterpretation."

• This quote conveys the speaker's desire for a clearer definition of AGI to set realistic expectations for its development timeline.

• The speaker is more concerned about the subtle, overlooked dangers of AGI than the widely discussed cataclysmic ones.
• There is a worry about society's ability to adapt to rapid changes brought on by AGI.
• The speaker highlights the importance of considering both known and unknown risks associated with AGI.

"I'm more concerned about the subtle dangers because I think we're more likely to overlook those...the unknown unknowns are really hard."

• The quote expresses the speaker's concern for the less obvious but potentially more insidious risks of AGI.

• The speaker believes that resilience is a teachable and increasingly important skill.
• There is an acknowledgment of the speaker's own potential biases and weaknesses, particularly a pro-technology bias.
• Self-awareness is discussed in the context of understanding one's motivations and the importance of adaptability.

"I believe resilience can be taught...and in the future, I think resilience and adaptability will be more important they've been in a very long time."

• This quote underscores the speaker's belief in the growing importance of resilience and adaptability in an ever-changing world.

"I think I have like a general bias to be too Pro technology just cuz I'm curious and I want to see where it goes."

• The quote reflects the speaker's self-awareness regarding their inclination towards technology and its potential impact on decision-making.

• The speaker is excited about GPT-5 being significantly smarter than its predecessors.
• GPT-5's intelligence is considered a remarkable fact in human history.
• The continuous improvement of GPT models is not confined to specific areas but is general.

"GPT 5 is going to be smarter than a lot smarter than GPT 4 GPT 6 going to be a lot smarter than GPT 5 and we are not near the top of this curve."

• This quote emphasizes the expected intelligence progression in GPT models, suggesting that each iteration will be substantially smarter than the last, and that there is still a significant potential for growth.

• The speaker is not worried about AI stifling human innovation, believing it will lead to more amazing things.
• Concerns are growing about responsible deployment as models become more capable.
• Strategies like red teaming, external audits, and iterative deployment with tight feedback loops are important.
• Negotiating AI rules with stakeholders is becoming increasingly complex.

"I really deeply believe that people will just surprise us on the upside with better tools."

• The speaker expresses confidence that providing people with advanced AI tools will lead to positive and surprising outcomes.

"I am though increasingly worried about how we're going to do this all responsibly."

• Despite the optimism for human innovation, there is a significant concern about responsibly managing the deployment of more capable AI models.

• Global and equitable access to AI compute resources is crucial.
• Training compute availability worldwide is a growing concern.
• Countries are recognizing the need for their own AI infrastructure, and the speaker hopes to assist in building this.

"I do think we get to a world where we sort of think about it as a human right to get access to a certain amount of compute."

• The speaker believes access to AI computing power could be considered a human right, highlighting the importance of global accessibility to AI resources.

• AI could be beneficial for space exploration due to the hostile environment for biological life.
• Sending robots to space is seen as easier than sending humans.

"I think space is like not that hospitable for biological life obviously. and so if we can send the robots that seems easier."

• The speaker suggests that AI and robotics are more suited for space exploration than humans, given the challenging conditions in space.

• Being contrarian and right is valuable, but being contrarian alone is not enough.
• It's important to trust one's intuition and thought process, which improves over time.
• Surrounding oneself with original thinkers and the right peer group can help, but part of the process must be done independently.

"The most important thing to be right and the group is usually right but where the most value is, is when you are contrarian and right."

• The speaker emphasizes the importance of not just having a unique idea but also ensuring that the idea is correct.

• Energy demand is expected to increase, correlating with quality of life.
• Fusion or solar plus storage are seen as the most likely dominant sources for electrical generation.
• Achieving one cent per kilowatt-hour energy is a goal for the future.

"My guess is that Fusion eventually dominates electrical generation on Earth."

• The speaker predicts that fusion energy will become the primary method for electricity generation on Earth due to its potential advantages.

• The speaker learned about the resilience and capability of OpenAI's team during a crisis.
• The decision to return to OpenAI was driven by a love for the organization and its mission.
• The speaker values the culture and aims to see the mission through with the team.

"I learned that the team was super resilient and like ready to kind of run the company."

• This quote reflects the speaker's realization that OpenAI's team was capable of managing the company even in his absence.

• OpenAI has a unique Russian doll structure with a nonprofit owning a for-profit entity.
• This structure was not planned from the beginning and evolved as the organization's direction changed.
• OpenAI did not initially plan to have a product, which influenced the eventual structure.

"If you're going to start a company you got to have like some theory that you're going to sell a product someday. and we didn't think we were going to have a product when we started."

• The speaker discusses the unexpected evolution of OpenAI's business model, which did not initially include plans for a product, influencing its organizational structure.

• The speaker, associated with OpenAI, emphasizes that the original intention for starting the organization was to advance AI research, not to make money or gain power.
• The speaker had already made significant money before starting OpenAI, indicating that financial gain was not a primary driver.
• OpenAI's structure, including a limited nonprofit board with fiduciary duties to humanity rather than shareholders, was designed to align with its mission but may evolve over time.
• The speaker believes in the importance of adapting organizational structure and mission as circumstances change.

"We just wanted to like push AI research forward... there's no motivation around making money or power... I had already made a lot of money."

• This quote highlights the speaker's assertion that OpenAI was founded with the primary goal of advancing AI research, rather than personal financial gain or power.

"If you look at the OpenAI structure... it's a nonprofit board which is supposed to look after its fiduciary duties to... Humanity."

• The quote explains that OpenAI's board is designed to prioritize the interests of humanity as a whole, reflecting the organization's broader ethical commitments.

"Life is not a problem set... you don't get to like solve everything really nicely all at once."

• The speaker reflects on the reality that real-life challenges, including those faced by organizations, are complex and cannot be solved in a straightforward manner like classroom problems.

• The speaker is uncertain about the specific changes AI will bring to geopolitics but believes it will be significant.
• There is an acknowledgment that AI could have a profound impact on the balance of power in the world, but the speaker is hesitant to predict exact outcomes.

"I have such a hard time saying what it's actually going to do... it feels like it does change a lot."

• This quote conveys the speaker's belief that AI will significantly alter geopolitics and the balance of power, even though they are unsure of the precise effects.

• The speaker warns against anthropomorphizing AI and suggests focusing on AI's ability to recognize its own errors and reasoning flaws.
• Building AI systems that can introspect and identify their own uncertainties is considered crucial for responsible AI deployment.

"How much can AI recognize its own flaws... that seems to me like really important to pursue."

• The speaker stresses the importance of developing AI that can self-assess and acknowledge its limitations, which is seen as a key aspect of responsible AI.

• The culture at OpenAI is driven by a shared mission and a deep sense of purpose.
• Employees tend to embrace the importance of the mission after joining, which strengthens the organization's culture.
• The speaker attributes the success of startups, including OpenAI, to a strong alignment with a binding mission.

"I think it really is like deep sense of purpose and loyalty to the mission."

• This quote emphasizes the speaker's belief that the strong culture at OpenAI is rooted in a collective commitment to its mission.

• The speaker acknowledges the potential for AI misuse and the importance of mitigating negative impacts while maximizing benefits.
• The speaker suggests a tight feedback loop between technology users and societal norms to balance safety and autonomy.
• The role of governments and democratic processes is highlighted as crucial in shaping the responsible use of AI.

"I believe that AI will be very net good... but I think like with any other tool... it'll be misused."

• The speaker recognizes that despite AI's potential for good, it is inevitable that it will also be misused, similar to any powerful tool.

"We're not going to get this perfectly right... a tight feedback loop I think is the best way to get it closest to right."

• The speaker accepts that society will not perfectly manage AI's impact, but advocates for a responsive approach to continually improve outcomes.

• The speaker compares the evolution of human capabilities to the potential of AI, suggesting that AI will contribute to societal "scaffolding" and enable future generations to achieve more.
• The notion that society, as a collective intelligence, has already surpassed individual human intelligence is used to frame the potential benefits of AI.
• The speaker expresses optimism that AI will be a net positive force, enabling new scientific discoveries and advancements.

"Society is far smarter than you now... Society is an AGI as far as you can tell."

• The speaker suggests that the collective intelligence of society, enabled by technology and infrastructure, already functions as a form of artificial general intelligence.

"Things that are smarter than us will contribute to that same scaffolding... your children will have tools available that you didn't."

• This quote reflects the speaker's optimistic view that AI will build upon existing societal knowledge and infrastructure, providing future generations with even more advanced tools and capabilities.