Below are 10 bold predictions about what will unfold in the world of artificial intelligence in 2021, from academic research to startups to capital markets to regulation. To keep ourselves honest, we will revisit these predictions in December 2021 to grade how we did.
1. Both Waymo and Cruise will debut on the public markets.
Autonomous vehicle developers like Waymo and Cruise have massive ongoing cash needs. Public market investors are thirsty for IPOs. The 2020 SPAC boom has provided a novel way for less mature businesses to go public. And SPAC investors have shown a voracious appetite for next-generation mobility companies (see, e.g., Nikola, Velodyne, Luminar, Innoviz, Canoo, Fisker, Romeo Systems).
Waymo and Cruise will take advantage of the market environment by going public in 2021. Fully spinning out from their parent companies—Alphabet and General Motors, respectively—will likely unlock significant enterprise value. Of the two, Waymo is more likely to pursue a traditional IPO.
2. A political deepfake will go mainstream in the U.S., fueling widespread confusion and misinformation.
Deepfake technology is rapidly improving and proliferating. Recent incidents in Gabon and Brazil reflect the technology’s destructive potential in the political sphere. 2021 will be the year that a piece of deepfake content goes mainstream in the United States, with a meaningful portion of the population initially believing it to be real. The deepfake will most likely be a video of a public figure making controversial comments.
In response, some policymakers will intensify calls to repeal Section 230 of the Communications Decency Act, arguing that big tech companies must be held responsible for policing the spread of deepfakes on their platforms.
3. The total number of academic research papers published on federated learning will surge.
Data privacy is becoming an increasingly urgent issue for consumers and regulators. Given this, privacy-preserving AI methods will continue to gain momentum as the most sustainable way to build machine learning models. The most prominent of these methods is federated learning.
The number of academic research papers published on federated learning has grown from 254 in 2018, to 1,340 in 2019, to 3,940 in 2020, according to Google Scholar. This exponential growth will continue: in 2021, over 10,000 research papers will be published on the topic of federated learning.
4. One of the leading AI chip startups will be acquired by a major semiconductor company for over $2B.
Silicon chips purpose-built for AI workloads are the future of the semiconductor industry. Intel’s $2 billion acquisition of Habana Labs last year was an acknowledgement of this reality. In 2021, in an effort to prevent itself from being disrupted, another legacy chipmaker will make a major acquisition of an AI chip startup.
Most likely acquisition targets: Graphcore, Cerebras, SambaNova
Most likely acquirors: NVIDIA, AMD, Qualcomm, Intel
5. One of the leading AI drug discovery startups will be acquired by a major pharmaceutical company for over $2B.
Big Pharma has woken up to the fact that machine learning offers the potential to revolutionize pharmaceutical drug discovery and development. In 2021, one of the major pharmaceutical companies will pay up to acquire an AI drug discovery startup, bringing its technology and talent in-house.
Most likely acquisition targets: Recursion, Exscientia, insitro, Atomwise
Most likely acquirors: Bayer, GlaxoSmithKline, Novartis, Bristol Myers Squibb, Eli Lilly, Gilead
6. The U.S. federal government will make AI a true policy priority for the first time.
The United States has lagged other countries, notably China, when it comes to proactive public policy support for artificial intelligence. This will begin to change in 2021 with a Biden White House and a more motivated Congress.
The Biden administration will put forth, and Congress will pass, a federal budget that dramatically increases government funding for AI. Congress will also pass into law a national strategy for AI that addresses topics like AI ethics, research priorities, national security implications and labor automation.
7. An NLP model with over one trillion parameters will be built.
In 2019 OpenAI published GPT-2, the first NLP model with over 1 billion parameters (it had 1.5 billion). At the time this was seen as staggeringly large. In 2020 OpenAI dropped GPT-3 on the world, which had a whopping 175 billion parameters.
The transformer “arms race” will continue in 2021 with the publication of the first model with over 1 trillion parameters. Most likely this model will come from OpenAI and be named GPT-4. Other organizations that might break the trillion-parameter-model mark include Microsoft, NVIDIA, Facebook and Google.
8. The “MLOps” category will begin to undergo significant market consolidation.
A spate of startups building tools and infrastructure for machine learning has emerged in recent years. Relatively few of these “AI picks and shovels” startups will survive as large standalone companies. Meaningful consolidation will begin to take place in this category in 2021.
Startups building specialized “point solutions” will be scooped up by larger players seeking to develop comprehensive, end-to-end model development platforms. Intel’s dual acquisitions of SigOpt and Cnvrg.io this year are canaries in the coalmine.
Likely acquisition targets: Alectio, Algorithmia, Arize AI, Arthur AI, Comet, DarwinAI, Fiddler Labs, Gradio, OctoML, Paperspace, Snorkel AI, Truera, Verta, Weights & Biases, et al.
Likely acquirors: IBM, Microsoft, Amazon, Databricks, DataRobot, Oracle
9. AI will become an important part of the narrative in regulators’ antitrust efforts against big tech companies.
Regulatory authorities in the U.S. and Europe formally initiated antitrust proceedings against Amazon, Apple, Facebook and Google this year. To this point, regulators have not explicitly focused on artificial intelligence as they have articulated the antitrust cases against the technology giants.
In the coming year, expect regulators and commentators to begin invoking AI more frequently as they frame how and why these companies are unfairly stifling competition. The core argument will be that the companies’ data monopolies give them insurmountable advantages in developing effective machine learning algorithms.
10. Biology will continue to gain momentum as the hottest, most transformative area to which to apply machine learning.
This is both the least measurable and the most important prediction on this list.
In terms of academic research, startup funding, and mainstream media attention, biology will increasingly emerge as the highest-impact, highest-consequence field to which to apply AI. DeepMind’s historic AlphaFold achievement last month—the ramifications of which will take years to fully play out—is merely a prelude to what humanity will accomplish by applying computational methods and machine learning to the mysteries of biology.
This article has been published from a wire agency feed without modifications to the text. Only the headline has been changed.
