2030,AIHow will the field change? Rob Toews, a venture capitalist at Radical Ventures, published 5 predictions about artificial intelligence in 2030.predict.
Here are five bold predictions about what the world of AI will look like in 2030. Whether you agree or disagree with them, I hope they’ll get you thinking.
Source: The image is generated by AI, and the image is authorized by Midjourney
1. Nvidia's market value will be significantly lower than it is now, and Intel's price will be significantly higher than it is now.
Nvidia is the hottest company in the world right now. It is the brainchild of the current generative AI craze.maximumThe beneficiaries have seen their market value soar from less than $300 billion at the end of 2022 to over $2 trillion today.
But Nvidia’s position as the single dominant supplier of AI chips cannot and will not last.
What Nvidia has built is difficult to replicate, but not impossible. A resurgent AMD is emerging as a credible alternative supplier of advanced GPUs.Cutting EdgeNvidia’s new MI300 chip is about to be widely available. Big tech companies — Amazon, Microsoft, Alphabet, Meta — are investing heavily in developing their own AI chips to reduce their reliance on Nvidia. OpenAI’s Sam Altman is seeking up to $1 trillion to build a new chip company to diversify the world’s supply of AI hardware.
As demand for AI chips continues to grow in the coming years, relentless market forces will ensure that more competitors enter, supply increases, prices fall, margins tighten, and Nvidia's market share declines.
In addition, as the market matures in the coming years, the main type of AI computing workload will shift from training to inference: that is, from building AI models to deploying them in real-world environments. Nvidia's highly specialized chips are unmatched in training models. But inference can be done with cheaper, more commoditized chips, which could weaken Nvidia's advantage in the market and create opportunities for competitors.
This isn’t to say that Nvidia won’t still be a vital part of the AI ecosystem in 2030. But the current surge in its stock price — which, as of this writing, makes it the world’s third-most valuable company — is nothing compared to Amazon or Alphabet — and in retrospect, it looks like irrational exuberance.
In the meantime: What separates Intel from nearly every other chip company in the world?
It makes its own chips.
Nvidia, AMD, Qualcomm, Broadcom, Alphabet, Microsoft, Amazon, Tesla, Cerebras, SambaNova, Groq: None of these companies make their own chips. Instead, they design chips and then rely on other companies—most importantly, TSMC—to manufacture them for them.
Intel solely owns and operates its own chip manufacturing facilities.
The ability to manufacture chips has become a key geopolitical asset. Case in point: China’s total reliance on foreign semiconductor suppliers has enabled the United States to hamper China’s domestic AI industry by banning imports of AI chips into China.
U.S. policymakers are acutely aware that chip manufacturing today is extremely concentrated inTaiwanThe vulnerability brought about by this, especially in ChinaTaiwanThe move comes as the Trump administration takes an increasingly tough stance. Promoting advanced semiconductor manufacturing in the United States has become a top policy priority for the U.S. government. U.S. lawmakers are taking decisive action to advance this goal, including committing up to $280 billion under the 2022 CHIPS Act.
Over the past decade, Intel hasCutting EdgeIt’s no secret that Intel lags behind TSMC in chip capabilities. Yet it remains one of the few companies in the world capable of making advanced semiconductors. Under CEO Pat Gelsinger, who took over in 2021, Intel has refocused its chipmaking priorities and pursued an ambitious strategy to recapture its former position as the world’s preeminent chipmaker. There have been recent signs that the company is making progress toward that goal.
Perhaps most importantly: as a chip manufacturer in the United StatesLeaders, there is no other choice.
U.S. Commerce Secretary Gina Raimondo, who is leading the Biden administration's AI and chip efforts, acknowledged this directly in a recent speech: "Intel is America's champion chip company. Simply put, America needs Intel. This bodes well for Intel's business prospects."
Nvidia’s current market cap is $2.2 trillion. Intel’s valuation is $186 billion, more than an order of magnitude smaller. We expect this gap to shrink significantly by 2030.
2. We will interact with various AIs in our daily lives as naturally as we interact with other humans today.
Although the whole world is talking about artificial intelligence, the common people today are stillCutting EdgeThe number of actual touchpoints an AI system has is limited: maybe an occasional query to ChatGPT or Google Bard/Gemini.
By 2030, this situation will change dramatically.
We will use AI as our personal assistants, mentors, career advisors, therapists, accountants, lawyers.
They will be ubiquitous in our working lives: performing analysis, writing code, building products, selling products, supporting customers, coordinating across teams and organizations, and making strategic decisions.
Yes, by 2030, it will be common for humans to have AI as significant others.
As with any new technology, there will be an adoption curve. Some segments of the population will adapt more easily to interacting with new AI companions; others will resist longer. The diffusion of AI in our society will be like Ernest Hemingway’s poem about how people go bankrupt.FamousThe lines unfold like this: "Gradually, and then suddenly."
But make no mistake: This transformation will be inevitable. It will be inevitable because AI will be able to do many of the things that humans do today, except cheaper, faster, and more reliably.
3. More than 100,000 humanoid robots will be deployed in the real world.
Today’s AI boom is taking place almost entirely in the digital realm.
Generative models that can hold an intellectual conversation on any topic, generate high-quality video on demand, or write complex code represent important advances in AI. But these advances are all happening in the world of software, in the world of bits. There is a whole field waiting to be explored today.Cutting EdgeChanges brought by artificial intelligence: the physical world, the atomic world.
Of course, the field of robotics has been around for decades, and today there are millions of robots operating around the world that can automate different types of physical activities.
But today’s robots have limited capabilities and limited intelligence. They are often designed for specific tasks, like moving boxes around a warehouse, or completing a specific step in a manufacturing process, or vacuuming the floor. They are far from having the fluid adaptability and generalizable understanding of large language models like ChatGPT.
This will change in the coming years. Generative AI will conquer the world of atoms — and it will make everything that has happened in AI so far seem trivial by comparison.
Dating back to the dawn of digital computing, a recurring theme in technology has been to make the hardware platform as general as possible and retain as much flexibility as possible for the software layer.
This principle was championed by the intellectual godfather of computing and artificial intelligence himself, Alan Turing, who immortalized it in the concept of the "Turing machine": a machine capable of executing any possible algorithm.
The early development of digital computers validated Turing’s fundamental insights. In the 1940s, different physical computers were built for different tasks: one to calculate the trajectory of a missile, another to decipher enemy messages. But by the 1950s, general-purpose, fully programmable computers had become the dominant computing architecture. Their versatility and adaptability across use cases proved to be a decisive advantage: they could be continually updated and used for any new application simply by writing new software.
In recent history, consider how many different physical devices have been broken down into one product, the iPhone, thanks to the genius of Steve Jobs and others: a phone, a camera, a video recorder, a voice recorder, an MP3 player, a GPS navigator, an e-book reader, a gaming device, a flashlight, a compass.
(A similar pattern can even be traced in the recent trajectory of AI models, though in this case everything is software. Narrow, function-specific models — one for language translation, another for sentiment analysis, and so on — have over the past few years given way to general-purpose “base models” that can perform a full range of downstream tasks.)
We will see the same shift in robotics over the next few years: away from specialized machines with narrowly defined use cases and towards more general-purpose hardware platforms that are flexible and adaptable.
What would this universal hardware platform look like? What form factor would it need to be flexible enough to operate in a variety of different physical environments?
The answer was clear: it needed to look human.
Our entire civilization has been designed and built by humans and for humans. Our physical infrastructure, our tools, our products, the size of our buildings, the size of our rooms, the size of our doors: all of it is optimized for the human body. If we want to develop a general-purpose robot that can operate in factories, warehouses, hospitals, stores, schools, hotels, and our homes, then that robot needs to be shaped like us. No other form factor can achieve the same effect.
That’s why the opportunity for humanoid robots is so huge.Cutting EdgeBringing AI into the real world is the next great frontier in AI.
Large language models will automate a lot of cognitive work in the coming years, while humanoid robots will automate a lot of manual work.
These robots are no longer a distant science fiction dream. Although most people don’t realize it yet, humanoid robots are about to be deployed in the real world.
Tesla is investing heavily in the development of a humanoid robot called Optimus. The company aims to start shipping the robot to customers in 2025.
Tesla CEO Elon Musk made no secret of how important he expects the technology to be to the company and the world: "I'm surprised that people don't realize the importance of the Optimus robotics program. The importance of Optimus will become obvious in the next few years. Those who have the insight or the ability to watch and listen carefully will understand that Optimus will ultimately be more valuable than Tesla's automotive business, more valuable than [Full Self-Driving]."
Some young startups are also making rapid progress here.
Just last week, Bay Area-based Figure announced a $675 million funding round from investors including Nvidia, Microsoft, OpenAI, and Jeff Bezos, and a few months ago released an impressive video of its humanoid robot making coffee.
Another leading humanoid startup, 1X Technologies, announced a $100 million funding round in January. 1X already offers one humanoid robot (with wheels) for sale and plans to release the next generation (with two legs) soon.
In the coming years, these companies will move from small-scale customer pilots to large-scale production, and by the end of the decade, we expect to see hundreds of thousands, if not millions, of humanoid robots deployed in the real world.
4. “Agents” and “AGI” will become outdated terms no longer in widespread use.
Two of the hottest topics in AI today are agents and artificial general intelligence (AGI).
Agents are AI systems that can complete loosely defined tasks: for example, planning and booking your upcoming travel. AGI refers to AI systems that meet or exceed human capabilities in all aspects.
When people look ahead to the state of AI in 2030, intelligent agents and/or AGI are often at the forefront.
Yet we predict that by 2030 neither term will even be in widespread use. Why? Because they will no longer be relevant as standalone concepts.
Let’s start with the Agent.
By 2030, agent behavior will become an essential element of any advanced AI system.
The umbrella term “agent” that we use today is really just a set of core capabilities that any truly intelligent entity possesses: the ability to think long-term, plan, and take action in pursuit of open-ended goals. Becoming “agent” is the natural and inevitable end state of today’s AI.Cutting EdgeAI systems don’t just generate output when prompted; they get the job done.
In other words, "agents" will no longer be an interesting subfield of AI research as it is today. AI will become agents, and agents will become AI. Therefore, the term "agent" as a standalone concept is useless.
How about the term “AGI”?
Artificial intelligence is fundamentally different from human intelligence, a basic fact that people often fail to grasp.
AI will become incredibly powerful in the coming years. But we will stop conceptualizing its trajectory as toward some “broad” end state, especially one whose contours are defined by human capabilities.
AI great Yann LeCun summed it up well: “There is no such thing as AGI…even humans are specialized.”
Using human intelligence as the ultimate anchor and measure of AI development fails to recognize all the powerful, profound, unexpected, socially beneficial, and completely non-human capabilities that machine intelligence may have.
By 2030, AI will be more powerful than humans, changing our world. It will also continue to lag behind human capabilities in other ways. If AI can understand and interpret every detail of human biology down to the atomic level, who cares if it has the “generality” to match human capabilities across the board?
The concept of general artificial intelligence is not particularly coherent. As AI rapidly advances in the coming years, the term will become increasingly unhelpful and irrelevant.
5. AI-driven unemployment will become one of the most widely discussed political and social issues.
Concerns about technology-induced job loss are a common theme in modern society, dating back to the Industrial Revolution and the Luddites. The age of AI is no exception.
But so far, much of the discussion about AI’s impact on the job market has been theoretical, long-term, and limited to academic studies and think tank white papers.
This will change much more abruptly than most people imagine. Before the end of this decade, AI-induced job loss will become a concrete and pressing reality in the lives of ordinary citizens.
We’re already starting to see the canaries in the coal mine here. Last month, fintech giant Klarna announced that its new customer service AI system was taking over the work of 700 full-time human agents. Plagiarism detection company Turnitin recently predicted that it would cut 20% of its workforce over the next 18 months due to advances in AI.
In the coming years, organizations will find that they can increase profitability and productivity by using AI to do more and more of the work that previously required humans. This will happen across industries and pay scales: from customer service agents to accountants, from data scientists to tellers, from lawyers to security guards, from court reporters to pathologists, from taxi drivers to management consultants, from journalists to musicians.
This isn't a far-fetched possibility. Today, the technology is good enough for many situations.
If we’re honest with ourselves, one of the main reasons we’re so excited about AI—one of the main reasons it offers such a transformative economic opportunity—is that it will be able to do things cheaper, faster, and with more accuracy than humans can today. Once AI can deliver on that promise, the need and economic rationale for hiring as many people as we do today in most fields will diminish. Almost by definition, in order for AI to have an impact on society and the economy, it will take people’s jobs away. New jobs will be created, of course, but not as quickly or in as many numbers, at least at first.
Job losses will cause great pain and disruption in the short term. Political movements andleaderThere will be strong opposition to this trend. Other parts of society will be equally strong in support of the benefits of technology and AI. Civil unrest and protests will be inevitable; they will undoubtedly become violent at times.
Citizens will call on their elected officials to act in a certain direction. Creative policy proposals like universal basic income will move from fringe theory to passed legislation.
There will be no easy solutions or clear moral choices. Political positions and social identities will increasingly be determined by one’s view of how society should guide the spread of AI throughout the economy.