Artificial Buildup: AI Startups Were Hot In 2023, But This Year May Be Slightly Different

Although venture investment continued to slow down in 2023, that was not the case for the AI sector.

Generative AI and AI-related startups raised nearly $50 billion in 2023, per Crunchbase data, with some, including OpenAI, Anthropic and Inflection AI, raking in billions of dollars all by themselves.

While investors have little doubt many AI startups will continue to push their valuations northward — especially those building their own models and platforms — the new year could provide a reckoning and recalibration for a market that seemed to know no bounds.

Big year

Before looking ahead, it may be important to take a step back. Although 2023 may be looked at as the “Year of OpenAI,” investors started to warm to the industry in late 2022, when London-based AI-driven visual art startup Stability AI, San Francisco-based AI video and audio editing tool Descript, and Austin, Texas-based AI content platform Jasper all raised big funding.

Then early last year, the round that shook the venture world — Microsoft’s reported $10 billion into AI titan OpenAI — set everything off and running. More than 70 rounds of $100 million or more went to startups creating models, providing infrastructure or applying the technology to a specific application in 2023.

That also included rounds outside the U.S., as Paris-based Mistral AI and Germany-based Aleph Alpha — both OpenAI competitors — raised rounds at or near $500 million.

The heat in the market caused valuations to spike.

Last May, Anthropic raised a $450 million Series C led by Spark Capital at a reported pre-investment valuation of $4.1 billion. Then in October, Anthropic reportedly raised up to $2 billion from previous investor Google. It was reported before the round that Anthropic was seeking a valuation between $20 billion and $30 billion.

Similarly, in early spring OpenAI picked up about $300 million from the likes of Sequoia Capital and Andreessen Horowitz at a valuation of around $29 billion. By the end of the year it was reportedly in talks to sell shares on the secondary market at a staggering $90 billion valuation (of course, some of the late-year drama slowed the process).

Hype cycle

Of course, those are two of the largest and most valuable startups in generative AI, and most investors would love to get in on them. However, other, less mature, startups in AI also saw valuations skyrocket, causing concern among investors that we may see some slowdown as 2024 wears on.

One of the biggest open questions is: Just how many winners can there be in AI?

Saad Siddiqui, a general partner at Telstra Ventures, said when he looks at the most recent big transition in tech — the mobile revolution — most of the large infrastructure winners were well-established tech players at the time — with startups such as Twilio being the exception.

The biggest startup winners came on the application layer — DoorDash, Uber, Instacart — which built mobile-native services to take advantage of the infrastructure being put in place, Siddiqui said.

“A similar trend could happen here, especially when you consider Big Tech has all the data,” he said.

Navin Chaddha, managing director at Menlo Park, California-based Mayfield Fund 1, echoed similar sentiments, saying certain layers of the AI stack like the semiconductor and cloud layer are already taken by big players. But some open spaces like middleware tools for developers and aspects of the application layer remain.

“Some layers are taken, but the rest of the layers are open,” he said.

However, AI is a resource-heavy endeavor. Startups need data, computing power, talent and deep pockets — all things Big Tech companies already have. While many of the big names are currently investing in AI startups, it will be interesting to see if they themselves aren’t the biggest winners of the AI revolution.

Regulation

Another area AI still has to grapple with will be legal and regulatory issues — which also could affect investment.

“You are seeing some startups working through some legal implications even now,” said Don Butler, managing director at Thomvest Ventures. “I think some of that will lead to a cooling off when it comes to investment, especially in early-stage AI.”

Concerns over regulatory issues hit center stage late in the year when the European Union finally agreed a provisional deal for AI rules that put a guardrail into place for the development of the technology — establishing a new standard others across the globe may follow.

Nearly at the same time, reports broke that the Federal Trade Commission is looking at Microsoft’s investment in OpenAI and whether it may violate antitrust laws — showing regulators were closely watching the dealings of Big Tech in the emerging market.

Money will flow

Of course, predicting a slight slowdown and recalibration in AI funding is not the same as a complete halt to investment in the space — which no one expects.

The industry already seems to have started a slowdown. As last year wore on, many investors appeared less and less interested in marketing or sales platforms that just wrapped AI around their platform, and news of AI startups having a hard time raising funding started to make headlines.

However, even at the end of the year startups like Essential AI and Vast Data were able to raise large rounds, proving inventor appetite still remained.

In fact, Essential AI’s raise is emblematic of what many AI startups have been able to do effectively when raising cash — attracting Big Tech to participate in the round. Essential’s round included participation from Advanced Micro Devices, Google and Nvidia.

While many of those big strategics can afford to invest at high valuations, the rising market could cause some VCs to reconsider — especially as venture firms have turned away from cash-burning startups and want to see actual solid financials from the companies they are funding.

Niko Bonatsos, a managing director at General Catalyst, said several AI startups definitely raised at inflated valuations and there likely already is regret among some investors.

However, Bonatsos also points out that as AI evolves — which it has been doing extremely rapidly — he expects some of the costs to come down.

“I think the next set of algorithms should be more efficient, so you’ll need less computer power,” he said. “Also, there will be more open source. So the cost of starting a company will go down.”

It will be intriguing to see if VCs can stop themselves from chasing some of the frothy valuations that have hit the space and, if they do, how startups in an expensive sector will react.

The new year brings a lot of questions — some even AI may not be able to answer.