Can AI Really Save the Future? A Lawyer’s Take on Sam Altman’s Optimistic Vision

In his latest essay, The Intelligence Age, Sam Altman envisions a future where ChatGPT and other AI technologies rapidly transform the world for the better. He projects that the exponential growth we’ve seen in generative AI will continue, unlocking astonishing advancements in science, society, and beyond. According to Altman, AI-driven breakthroughs could soon lead us into a virtual utopia—one where the possibilities seem limitless.

A Balanced Approach

Unlike many futurists, Sam Altman, in his essay The Intelligence Age, only briefly touches on the potential problems and dangers ahead. While this overwhelmingly optimistic vision may come across as a sales pitch, to his credit, there’s more depth to it. Altman’s predictions are rooted in science and insider knowledge, which means his ideas should be taken seriously—but with a healthy dose of skepticism. In this article, I’ll provide that necessary skepticism, highlighting the overlooked risks and exploring the darker side of AI. This balanced view will help you critically assess Altman’s predictions and form your own conclusions.

Safe Predictions

Sam Altman begins The Intelligence Age by providing a time context for all of the predictions he will make: the next couple of decades. These are things he expects to happen by 2044. That is smart of Sam to put such a long time frame on his predictions. If they do not come true, then no one will remember. And if per happy chance they do come true, and so we are living in a near utopia, we will be happy for him to remind us that he called it back in 2024! And no doubt we will be happy to have our AIs build yet another statute in his honor!

Images by Ralph Losey using his Visual Muse GPT.

Sam continues his essay opening by saying that by 2044, “we will be able to do things that would have seemed like magic to our grandparents.” That is a safe prediction, which, as both a grandparent and a big dreamer, I can say with authority. We are already doing things most grandparents could never have dreamed about. So, the opening paragraph is attention grabbing, but really does not say much, which is often the way with ChatGPT.

The second paragraph is just two sentences of the typical, predictive fluff that we see from LLMs.

This phenomenon is not new, but it will be newly accelerated. People have become dramatically more capable over time; we can already accomplish things now that our predecessors would have believed to be impossible.

Sam Altman, The Intelligence Age.

Sam here basically repeats what he has already said in the first paragraph, but at least doesn’t say grandparents again. Starting to get bored? Me too. So let’s move onto the third paragraph:

We are more capable not because of genetic change, but because we benefit from the infrastructure of society being way smarter and more capable than any one of us; in an important sense, society itself is a form of advanced intelligence. Our grandparents – and the generations that came before them – built and achieved great things. They contributed to the scaffolding of human progress that we all benefit from. AI will give people tools to solve hard problems and help us add new struts to that scaffolding that we couldn’t have figured out on our own. The story of progress will continue, and our children will be able to do things we can’t.

Sam Altman, The Intelligence Age.

Now he expands on the same theme of rapid change and the totally stoked grandparents again. He compares slow, individual, physical evolution based on DNA to rapid, social cultural evolution based on AI. Sam refers to the resources of virtual experts and personal tutors. This is a good start to setting the stage of actual predictions.

Leading With Well Supported Predictions of AI Experts and Virtual Tutors

Now moving on to the fourth paragraph by Sam Altman, who is, in full disclosure, a person I admire. He is a longtime meditator who takes action and gets amazing things done. In this paragraph Sam’s real predictions begin:

It won’t happen all at once, but we’ll soon be able to work with AI that helps us accomplish much more than we ever could without AI; eventually we can each have a personal AI team, full of virtual experts in different areas, working together to create almost anything we can imagine. Our children will have virtual tutors who can provide personalized instruction in any subject, in any language, and at whatever pace they need. We can imagine similar ideas for better healthcare, the ability to create any kind of software someone can imagine, and much more.

Sam Altman, The Intelligence Age.

AI experts and tutors are already among us and used every day by millions. I personally think it is ChatGPT’s greatest achievement so far and have written extensively about it.

• ChatGPT’s Surprising Ability to Split into Multiple Virtual Entities to Debate and Solve Legal Issues
• Panel of AI Experts for Lawyers
• Panel of Experts for Everyone About Anything
• BACK TO SCHOOL: A Review of Salman Khan’s New Book, ‘Brave New Words: How AI will revolutionize education (and why that’s a good thing)’

As I am totally aligned with these opening predictions, I do not have much skepticism to add to Sam’s fourth paragraph. Well, one thing I could add for balance, which is something that educational experts like Sal Khan who promote the smart use of ChatGPT are always careful to point out, namely that human educators offer emotional support and mentorship that AI cannot provide. Interpersonal human relationships in class should have more time to flourish in the AI home tutor future, not less.

In a CLE class that I just taught with two federal judges as my co-teachers, I asked ChatGPT Omni to try to respond to a question that the District Court judge had just answered on the importance of human interaction for young law students. He did not want young lawyers to get too tied up with AI on their phones and stressed the importance of human interaction for real learning. The AI did not hear the judge’s answer, or know anything about the judge’s attitude, and yet the response the AI gave was eerily similar.

LAWYER QUESTION: Would you give us your opinion as to the importance of human interaction for young law students in learning the law today and learning how to use AI in the law?

ChatGPT RESPONSE: Human interaction is essential for young law students, particularly in learning the intricacies of law and how to use AI effectively. Law isn’t just about knowing statutes or cases; it’s about understanding human behavior, ethics, advocacy, and the nuances of judgment—all of which come from direct engagement with mentors, peers, and real-world experiences. AI can be a powerful tool, but it requires human discernment to interpret, apply, and contextualize its outputs. By learning from experienced legal professionals and collaborating with others, law students develop the critical thinking, ethical grounding, and communication skills necessary to wield AI responsibly and effectively in their future practice.

Everyone was taken aback upon hearing this, even the District Court Judge who had been somewhat suspect of AI until then. Nothing like unexpectedly hearing an AI think and speak just like you. So, I must say the first appearance of an AI in Court in Orlando was very successful. I predict it will not be its last.

Economic Prosperity

Now for Sam Altman’s next paragraph, an important one on economic prosperity.

With these new abilities, we can have shared prosperity to a degree that seems unimaginable today; in the future, everyone’s lives can be better than anyone’s life is now. Prosperity alone doesn’t necessarily make people happy – there are plenty of miserable rich people – but it would meaningfully improve the lives of people around the world.

Sam Altman, The Intelligence Age.

The unmentioned risk here is that AI, if not carefully regulated, could deepen the economic divide that already exists between rich and poor. According to the Federal Reserve Board in the second quarter of 2024, the top 1% of households in the U.S. held 30.2% of the country’s wealth and the top 10% held 66.7%. The Fed – Table: Distribution of Household Wealth in the U.S. since 1989. The Federal Reserve Board also reports that the bottom 50% of the U.S. population held only 2.5% of the country’s wealth.

Further wealth concentration among a few families, tech companies and countries who control AI resources is a real risk. Without international cooperation and regulations, Sam’s imagined utopias could be a luxury for only a select few, and AI could further increase global inequality rather than alleviate it.

Still, it is a remarkable achievement that so far ChatGPT and other generative AIs have been available to all, either free or at an affordable price. It has not become a tool for an elite few. In fact the rapid proliferation of generative AI so far is incredible. A recent broad spectrum survey in the U.S. with 5,014 responses indicates that 39% adults aged 18-64 have used generative AI. Moreover, 28% of employees use it regularly, and 10.6% use it daily. Bick, Blandin & Deming, The Rapid Adoption of Generative AI (National Bureau of Economic Research, Sept. 2024).

Historical Perspective and Superintelligence

The next two paragraphs seem to have Sam’s personal touch and poetic quality. I especially like the reference at the start to silicon as melted sand.

Here is one narrow way to look at human history: after thousands of years of compounding scientific discovery and technological progress, we have figured out how to melt sand, add some impurities, arrange it with astonishing precision at extraordinarily tiny scale into computer chips, run energy through it, and end up with systems capable of creating increasingly capable artificial intelligence.

This may turn out to be the most consequential fact about all of history so far. It is possible that we will have superintelligence in a few thousand days (!); it may take longer, but I’m confident we’ll get there.

Sam Altman, The Intelligence Age.

I agree about the most important invention in human history, which I explained in detail recently. Artificial General Intelligence, If Attained, Will Be the Greatest Invention of All Time (8/12/24). Also see Ray Kurzweil’s New Book: The Singularity is Nearer (when we merge with AI) (7/17/24).

To keep it real, however, all of the past technology shifts were buffered by incremental growth. AI’s exponential acceleration could overwhelm our existing systems. We need to try to manage this transition and to do that we need to consider the problems that come with it. See Seven Problems of AI: an incomplete list with risk avoidance strategies and help from “The Dude” (8/6/24).

Bigger Is Better: How Generative AI Came to Pass

Now we get into Sam Altman’s concise, well written summary of how LLM’s like ChatGPT work and evolved so quickly.

How did we get to the doorstep of the next leap in prosperity?
In three words: deep learning worked.
In 15 words: deep learning worked, got predictably better with scale, and we dedicated increasing resources to it.

That’s really it; humanity discovered an algorithm that could really, truly learn any distribution of data (or really, the underlying “rules” that produce any distribution of data). To a shocking degree of precision, the more compute and data available, the better it gets at helping people solve hard problems. I find that no matter how much time I spend thinking about this, I can never really internalize how consequential it is.

There are a lot of details we still have to figure out, but it’s a mistake to get distracted by any particular challenge. Deep learning works, and we will solve the remaining problems. We can say a lot of things about what may happen next, but the main one is that AI is going to get better with scale, and that will lead to meaningful improvements to the lives of people around the world.

Sam Altman, The Intelligence Age.

That is a standard Neural scaling law explanation of how deep learning works. It is one that OpenAI and most other LLM generative AI companies provide. There is good science behind it. See foundational scientific paper by Jared Kaplan, Sam McCandlish, et al, Scaling Laws for Neural Language Models (2020). Bigger is better, or as AI scientists put it, exponential increases in the scale of data and compute. This increase is what is what OpenAI reports caused the dramatic increase in intelligence in their ever larger ChatGPT models 3.5 and 4.0. The lines on all graphs are now moving almost straight up, which as mentioned, is nothing like the slower linear growth we have seen before.

An Epoch AI research report states that scaling can continue at least until 2030 when it can attain a 10,000-fold scale-up relative to current models. Jason Dorrier, AI Models Scaled Up 10,000x Are Possible by 2030, Report Says (Singularity Hub, 8/29/24). Just think of that, a scaling increase ten thousand times larger. Constraints on further increases are predicted by Epoch AI after 2030, primarily from power consumption limits. According to the Singularity Hub article Microsoft and OpenAI are now working together to raise unprecedented amount of funding for the next six years.

But spending will need to grow even more. Anthropic CEO Dario Amodei estimates models trained today can cost up to $1 billion, next year’s models may near $10 billion, and costs per model could hit $100 billion in the years thereafter. That’s a dizzying number, but it’s a price tag companies may be willing to pay. Microsoft is already reportedly committing that much to its Stargate AI supercomputer, a joint project with OpenAI due out in 2028. (hyperlink added)

Jason Dorrier, AI Models Scaled Up 10,000x Are Possible by 2030, Report Says.

The scaling continues to work, but has environmental costs and other detrimental effects not mentioned here by Sam (who is a big investor in nuclear energy). How to manage AI’s energy demand (World Economic Forum, 4/25/24). Also, companies are now discovering that there may be better ways than scaling to increase LLM intelligence going forward, primarily by improving the quality of the data and efficiency of the compute. See eg. Are bigger language models always better? (IBM, 7/15/24).

There are certainly large increases in carbon pollution caused by the generation of power required for big compute. See e.g. AI and energy: Will AI help reduce emissions or increase demand? (World Economic Forum, 7/22/24); A.I. Could Soon Need as Much Electricity as an Entire Country (NYT, 10/10/23). Demand Growth Offers Opportunities for Data Centers (The Current 3/19/24) (data centers’ total electricity consumption could almost double from 2022 to 1,000 TWh in 2026. That is equal to the entire electricity consumption of Japan).

NVIDIA claims it is making good progress in improving the quality of compute and reducing environmental impacts by invention (with AI help) of its latest “Blackwell” superchips. NVIDIA to Present Innovations at Hot Chips That Boost Data Center Performance and Energy Efficiency (NVIDIA,); Sustainable Strides: How AI and Accelerated Computing Are Driving Energy Efficiency (NVIDIA, 7/22/24). This could just be hype, but I don’t think so. All major chip makers now have that efficiency goal. See e.g. Beth Kindig, AI Power Consumption: Rapidly Becoming Mission-Critical (Forbes Newsletter, 6/20/24).

Predictions of Personal Assistants and Scientific Progress

Now Sam Altman goes on to state his positive vision of the future.

AI models will soon serve as autonomous personal assistants who carry out specific tasks on our behalf like coordinating medical care on your behalf. At some point further down the road, AI systems are going to get so good that they help us make better next-generation systems and make scientific progress across the board.

Sam Altman, The Intelligence Age.

I agree with Sam to a point, but he does not mention the limitations of AI. To understand these limitations you need to make a clear distinction between human and machine cognitive processes. The human mind is deeply linked and arises out of bodily experiences and the external world, whereas computational models lack a real world, experiential basis. As lawyers we must recognize the limits of mere machine tools. We cannot over-delegate to them just because their language sounds good, especially when we are serving as legal counselors, judges, and mediators. See e.g. Yann LeCun and Browning, AI And The Limits Of Language (Noema, 8/23/22) (“An artificial intelligence system trained on words and sentences alone will never approximate human understanding.”); Valmeekam, et al, On the Planning Abilities of Large Language Models (arXiv, 2/13/23) (poor at planning capabilities); Dissociating language and thought in large language models (arXiv, 3/23/24) (poor at functional competence tasks).

As I always say, trust but verify. See e.g. White House Obtains Commitments to Regulation of Generative AI from OpenAI, Amazon, Anthropic, Google, Inflection, Meta and Microsoft (8/1/23); Worrying About Sycophantism (7/9/24).

Sam’s Vision of Astounding Triumphs Ahead

Back to the next five paragraphs of Sam Altman’s feel-good essay.

Technology brought us from the Stone Age to the Agricultural Age and then to the Industrial Age. From here, the path to the Intelligence Age is paved with compute, energy, and human will.

If we want to put AI into the hands of as many people as possible, we need to drive down the cost of compute and make it abundant (which requires lots of energy and chips). If we don’t build enough infrastructure, AI will be a very limited resource that wars get fought over and that becomes mostly a tool for rich people.

We need to act wisely but with conviction. The dawn of the Intelligence Age is a momentous development with very complex and extremely high-stakes challenges. It will not be an entirely positive story, but the upside is so tremendous that we owe it to ourselves, and the future, to figure out how to navigate the risks in front of us.

I believe the future is going to be so bright that no one can do it justice by trying to write about it now; a defining characteristic of the Intelligence Age will be massive prosperity.

Although it will happen incrementally, astounding triumphs – fixing the climate, establishing a space colony, and the discovery of all of physics – will eventually become commonplace. With nearly-limitless intelligence and abundant energy – the ability to generate great ideas, and the ability to make them happen – we can do quite a lot.

Sam Altman, The Intelligence Age.

Now here are my spoiler comments. It is critical to keep generative AI inexpensive and democratize AI. But there are serious geopolitical risks involved, especially if we fail at that. Sam refers to “very complex and extremely high-stakes challenges” but does not mention the geopolitical risks. Nations with access to massive computational infrastructure could dominate AI development, leading to new forms of digital colonialism. This uneven distribution could trigger conflicts over technology access, much like the resource wars of the past. For instance, China could invade Taiwan for its AI chip manufacturing plants, that no one else has, even the U.S. How Taiwan’s Semiconductor Industry Prepares for a Potential Chinese Invasion (Manufacturing Today, 5/30/24). Remember that both NVIDIA and AMD just design the chips and outsource the chip manufacturing. What you need to know about Nvidia and the AI chip arms race (Marketplace, 5/8/24).

Sam Altman also does not address the accountability mechanisms that everyone agrees, Sam included, are needed to manage the complex systems he envisions. Without effective ethical frameworks and international governance, AI’s risks—misuse in warfare, surveillance, or economic manipulation—could spiral out of control. It’s not enough to simply trust in technological progress. We must also build legal and ethical guardrails. We must verify.

Altman’s prediction of AI solving monumental global challenges, from climate change to space exploration, sounds utopian. While AI can undoubtedly assist in these areas, it is naive to think that technology alone will overcome political, economic, and social barriers. Climate change, for instance, is as much a governance and resource issue as it is a technological one. Relying solely on AI to save the world without addressing human-made political challenges is a recipe for disappointment, perhaps even disaster. The technology is powerful, but it is no magic bullet. It is a mix between a Pandora’s box and a wish fulfilling Genie, having both good and bad potentials.

Sam’s Conclusion to The Intelligence Age

Back to Sam Altman, we reach the eloquent close of his essay where he finally gives lip service to the problems.

As we have seen with other technologies, there will also be downsides, and we need to start working now to maximize AI’s benefits while minimizing its harms. As one example, we expect that this technology can cause a significant change in labor markets (good and bad) in the coming years, but most jobs will change more slowly than most people think, and I have no fear that we’ll run out of things to do (even if they don’t look like “real jobs” to us today). People have an innate desire to create and to be useful to each other, and AI will allow us to amplify our own abilities like never before. As a society, we will be back in an expanding world, and we can again focus on playing positive-sum games.

Many of the jobs we do today would have looked like trifling wastes of time to people a few hundred years ago, but nobody is looking back at the past, wishing they were a lamplighter. If a lamplighter could see the world today, he would think the prosperity all around him was unimaginable. And if we could fast-forward a hundred years from today, the prosperity all around us would feel just as unimaginable.

Sam Altman, The Intelligence Age.

I happen to agree with him Sam on the jobs issue and consider it a straw man type argument allowing him to dodge more difficult issues. Also see my article pertaining indirectly to this topic, Sam Altman’s Favorite Unasked Question: What Will We Do in the Future After AI? (7/7/23).

All of history easily rebuts the notion that there will be no work left for humans. The Original Luddites Raged Against the Machine of the Industrial Revolution (History Channel, 1/4/19). AI will create entirely new jobs. For instance for lawyers, new jobs pertaining to AI regulations are emerging. AI will also change existing jobs for the better. It is already replacing the most boring parts of our work, leaving us to focus on the more rewarding and human aspects. Moreover, it is true that no worker will be replaced by an AI, they will be replaced by a human that knows how to use AI.

There is room for debate on both sides of course, and I realize this is a sincere concern of many, which is why many of us have addressed this issue at length. What Lawyers Think About AI, Creativity and Job Security (7/28/23); Jensen Huang’s Life and Company – NVIDIA: building supercomputers today for tomorrow’s AI, his prediction of AGI by 2028 and his thoughts on AI safety, prosperity and new jobs. (12/18/23); McKinsey Predicts Generative AI Will Create More Employment and Add 4.4 Trillion Dollars to the Economy2023 (12/18/23).

Sam could have addressed many even darker problems than unemployment. That would have destroyed the tone of his intentionally feel good article. Some of these issues have already been addressed by me here, but there are many more, including:

• Bias in the data the AI was trained on still remains and much more effort must be put into correcting that. Most of us in the field have written extensively on this important subject. See e.g. a few of my articles, each of which includes many citations to the writings of others. Stochastic Parrots: the hidden bias of large language model AI (3/25/24); Navigating the High Seas of AI: Ethical Dilemmas in the Age of Stochastic Parrots (3/4/24).
• Seven Problems of AI (8/6/24):
  (1) risks of AI in terms of privacy and data security;
  (2) exacerbation of existing biases and inequalities in society;
  (3) AI decision-making in critical areas like healthcare and criminal justice;
  (4) misuse of AI for malicious purposes, such as in cyberattacks or deepfakes;
  (5) lack of transparency and accountability;
  (6) protection against infringe on human rights, such as persecution of minorities; and,
  (7) dangers of relying too heavily on AI for decision-making processes, including by the military.
• Legal evidence problems created by AI, including deepfakes, such that we need to revise the rules now. The Problem of Deepfakes and AI-Generated Evidence: Is it time to revise the rules of evidence? Part One and Part Two.
• Manipulative empathic AI. Code of Ethics for “Empathetic” Generative AI (7/12/23) review of proposal by Jon Neiditz.
• Host of other ethics related issues, including regulatory. See the separate website I maintain on AI Ethics. It outlines many of the problems raised by AI and includes many links to groups working on them and videos on the topic. Also See: AI Ethics Website Updated (8/7/23).

Conclusion

Sam Altman’s essay, The Intelligence Age, offers a compelling vision of how AI can dramatically improve life in the coming decades. I encourage you to read it in its entirety, without interruption, to truly grasp the uplifting future he envisions.

However, as captivating as Altman’s optimism is, we must balance it with a dose of realism. The road ahead is filled with hurdles—from economic inequality to ethical concerns and geopolitical tensions—that cannot be ignored. These challenges should not dampen our hope but inspire us to shape AI’s development thoughtfully and responsibly. If we are serious about realizing this potential, we need to approach AI with both excitement and caution, ensuring that the benefits are shared by all.

As AI continues to evolve, the decisions we make today will determine its impact on humanity tomorrow. Now is the time to engage—whether you are a lawyer, technologist, or concerned citizen—and ensure that AI is developed with transparency, fairness, and ethical integrity. The future Altman envisions can be ours, but only if we actively work to make it a reality. Let’s meet that challenge together.