AI
Prediction Consensus: What the Experts See Coming in 2023
Prediction Consensus: What the Experts See Coming in 2023
In this, our fourth year of Prediction Consensus (now part of our more comprehensive 2023 Global Forecast Series), we’ve learned a few things about the universe of predictions, experts, outlooks, and forecasts.
- Experts are reasonably good at predicting the future one year out, though they are also in a strong position to help shape the future through their influential thought leadership and actions.
- Situations can and will flare up in unexpected ways, which can have knock-on effects on the whole system (e.g. COVID-19, Ukraine invasion).
- Experts are just as susceptible to hype as the rest of us, as evidenced by the glut of Web3 predictions in 2022 and AI predictions this year.
Of course, we’re susceptible to hype as well, which is why we asked ChatGPT to write the intro to this article:
Not bad. But, simple curiosity aside, it’s the practical considerations we’ll focus on today. This article serves as an overview of how experts think the markets will move, how trends will develop, and which risks and opportunities to watch over the coming 12 months.
Let’s gaze into the crystal ball.
The Economic Vibe Check
First, we’ll look at some big picture themes, and how experts see them playing out over 2023.
Inflation: This was the top economic story of last year, so it’s a natural starting place. Many of the expert opinions in this year’s database (now at 500+ predictions) are pointing to inflation easing off as the year progresses*. On the downside, few predict that inflation will drop back down to the 2% range that Fed policymakers favor.
GDP: Forecasters have been revising their economic projections downward in recent weeks. The latest was World Bank, which now sees global growth declining to 1.7% in 2023, down from 3% just six months ago. Most of the predictions in our database see global economic growth in the range of 1.5% to 2%.
Recession: As 2022 came to a close, the broad sentiment among experts in the financial industry is that recession is all but inevitable in developed markets this year. As dawn breaks in 2023, a few analysts now feel that the U.S.—and possibly Europe—could narrowly avoid recession.
Markets: Experts on Wall Street and beyond are cautiously optimistic about equities, and after the worst year on record for bonds in 2022, most analysts are declaring that “Bonds are back”.
*Interestingly, this was also last year’s prediction, but the scale of Russia’s invasion of Ukraine was a curve ball that caught many experts off guard.
AI is Eating the World
Jobs being displaced by automation is far from a new theme, but given the exponential improvements in AI in recent years, the risk to entire industries feels more existential today.
As an example, let’s consider art and design. One of the ways many illustrators and artists earn a living is through commissions—essentially being hired and paid to create a specific piece of art in their style.
Today though, free, powerful AI tools, such as Midjourney, allow users to generate high-quality art in an infinite number of styles with just a few clicks. Real art will never truly go out of style, and accomplished artists will always attract an audience, but this one example shows how quickly technology can disrupt an industry. (Artists can take solace in the fact that AI is still comically bad at rendering hands.)
Of course, there are obvious positive aspects to this technological advancement as well. Generative AI tools are useful for generating ideas and mock-ups, and even functional snippets of code. AI systems like AlphaFold unlock a world of possibilities in scientific domains.
From the hundreds of predictions we evaluated, it’s clear that experts view AI as a major catalyst this year. AI start-ups are forcing Big Tech to innovate faster, and employees are finding new ways to use AI-powered tools to increase productivity.
Experts predict that AI will impact peoples’ lives in a much more visible and tangible way in 2023 than in past years.
The China Factor
As world’s second largest economy and linchpin of global trade, events in China have a major impact on the world economy.
Xi Jinping’s reversal of Zero-COVID restrictions should drastically change the trajectory of the country’s economy. For one, reopening will unleash a flood of household spending and consumption.
China’s reopening will also impact other economies as well. For example, the resumption of travel will be a boon to destinations favored by Chinese vacationers. Economically, Hong Kong stands to benefit immensely—its GDP could jump upwards of 8% after reopening is complete. Emerging market commodity exporters could see a lift as well, though inflation could be reinvigorated as a result.
In the U.S., a storm is brewing over the extremely popular video app, TikTok. Many experts predict that regulators will either ban the app altogether in 2023, or force the sale of the company to an American entity. Regardless how that situation plays out, it underscores the souring relationship between the U.S. and China. The rivalry will continue to have ripple effects on the global markets throughout the year.
Energy
Energy was the S&P 500’s top performing sector two years in a row, and many experts feel that more growth is on the horizon.
The global system that supplies us with energy is breathtakingly complex, with a lot of unpredictable factors at play. Of all factors, conflict can create the most volatility, and 2023 has a number of geopolitical risks that could impact energy supplies. First, Europe will continue to diversify its energy imports away from Russia. Recently, liquefied natural gas from the U.S. has helped fill gaps.
Next, Iran could be a flashpoint in the Middle East this year. A brewing conflict in the region could cause instability, which will have knock-on effects on the energy industry—particularly in the event of attacks on oil and gas infrastructure.
Here are a few other factors to consider this coming year:
- The U.S. Energy Department will aim to replenish its Strategic Petroleum Reserve
- Easing of U.S. sanctions on Venezuela could lay the ground work for increased oil production
- In post-Zero-COVID China, economic activity will increase, pushing up demand
- In the UK, the energy price guarantee will rise in April, meaning higher energy bills for households
The Elon Playbook
After a lull in December (nobody wants to be the company that fires people during the holiday season) tech and tech-adjacent companies have resumed their zealous slashing of headcounts.
There had been a slew of layoffs already in 2023, topped by Salesforce, which is trimming 7,000 jobs, and Amazon, which is cutting 18,000 roles—primarily impacting the corporate side of the business.
Given the influence of Elon Musk in the tech industry, many experts are suggesting that his strategy of ruthlessly slashing headcount at Twitter might serve as inspiration for other technology leaders.
Employees in the tech industry are very well compensated, and many were hired during periods of intense competition between companies to attract talent and capture market share.
During a downturn, it’s tempting—and often necessary—for companies to course-correct. There were also predictions that the whole start-up and investment ecosystem could be switching from a hypergrowth to a value-focused mindset, which is a theme that is worth consideration in 2023.
AI
Charted: The Exponential Growth in AI Computation
In eight decades, artificial intelligence has moved from purview of science fiction to reality. Here’s a quick history of AI computation.

Charted: The Exponential Growth in AI Computation
Electronic computers had barely been around for a decade in the 1940s, before experiments with AI began. Now we have AI models that can write poetry and generate images from textual prompts. But what’s led to such exponential growth in such a short time?
This chart from Our World in Data tracks the history of AI through the amount of computation power used to train an AI model, using data from Epoch AI.
The Three Eras of AI Computation
In the 1950s, American mathematician Claude Shannon trained a robotic mouse called Theseus to navigate a maze and remember its course—the first apparent artificial learning of any kind.
Theseus was built on 40 floating point operations (FLOPs), a unit of measurement used to count the number of basic arithmetic operations (addition, subtraction, multiplication, or division) that a computer or processor can perform in one second.
Computation power, availability of training data, and algorithms are the three main ingredients to AI progress. And for the first few decades of AI advances, compute, which is the computational power needed to train an AI model, grew according to Moore’s Law.
Period | Era | Compute Doubling |
---|---|---|
1950–2010 | Pre-Deep Learning | 18–24 months |
2010–2016 | Deep Learning | 5–7 months |
2016–2022 | Large-scale models | 11 months |
Source: “Compute Trends Across Three Eras of Machine Learning” by Sevilla et. al, 2022.
However, at the start of the Deep Learning Era, heralded by AlexNet (an image recognition AI) in 2012, that doubling timeframe shortened considerably to six months, as researchers invested more in computation and processors.
With the emergence of AlphaGo in 2015—a computer program that beat a human professional Go player—researchers have identified a third era: that of the large-scale AI models whose computation needs dwarf all previous AI systems.
Predicting AI Computation Progress
Looking back at the only the last decade itself, compute has grown so tremendously it’s difficult to comprehend.
For example, the compute used to train Minerva, an AI which can solve complex math problems, is nearly 6 million times that which was used to train AlexNet 10 years ago.
Here’s a list of important AI models through history and the amount of compute used to train them.
AI | Year | FLOPs |
---|---|---|
Theseus | 1950 | 40 |
Perceptron Mark I | 1957–58 | 695,000 |
Neocognitron | 1980 | 228 million |
NetTalk | 1987 | 81 billion |
TD-Gammon | 1992 | 18 trillion |
NPLM | 2003 | 1.1 petaFLOPs |
AlexNet | 2012 | 470 petaFLOPs |
AlphaGo | 2016 | 1.9 million petaFLOPs |
GPT-3 | 2020 | 314 million petaFLOPs |
Minerva | 2022 | 2.7 billion petaFLOPs |
Note: One petaFLOP = one quadrillion FLOPs. Source: “Compute Trends Across Three Eras of Machine Learning” by Sevilla et. al, 2022.
The result of this growth in computation, along with the availability of massive data sets and better algorithms, has yielded a lot of AI progress in seemingly very little time. Now AI doesn’t just match, but also beats human performance in many areas.
It’s difficult to say if the same pace of computation growth will be maintained. Large-scale models require increasingly more compute power to train, and if computation doesn’t continue to ramp up it could slow down progress. Exhausting all the data currently available for training AI models could also impede the development and implementation of new models.
However with all the funding poured into AI recently, perhaps more breakthroughs are around the corner—like matching the computation power of the human brain.
Where Does This Data Come From?
Source: “Compute Trends Across Three Eras of Machine Learning” by Sevilla et. al, 2022.
Note: The time estimated to for computation to double can vary depending on different research attempts, including Amodei and Hernandez (2018) and Lyzhov (2021). This article is based on our source’s findings. Please see their full paper for further details. Furthermore, the authors are cognizant of the framing concerns with deeming an AI model “regular-sized” or “large-sized” and said further research is needed in the area.
Methodology: The authors of the paper used two methods to determine the amount of compute used to train AI Models: counting the number of operations and tracking GPU time. Both approaches have drawbacks, namely: a lack of transparency with training processes and severe complexity as ML models grow.
-
Energy2 weeks ago
What Electricity Sources Power the World?
-
Money3 weeks ago
Mapped: The Richest Billionaires in U.S. States
-
Markets2 weeks ago
The 25 Worst Stocks by Shareholder Wealth Losses (1926-2022)
-
Jobs3 weeks ago
Charted: Youth Unemployment in the OECD and China
-
Technology1 week ago
Visualizing Google’s Search Engine Market Share
-
Markets3 weeks ago
The Monthly Cost of Buying vs. Renting a House in America
-
Money1 week ago
Visualized: How Long Does it Take to Double Your Money?
-
Sports3 weeks ago
Ranked: The World’s Largest Stadiums