Business Reporter – AI and Automation

AI For Business

In an age of investment boom around generative AI, where many companies are overvalued and overinvested, Northern Data's Rosanne Kincaid-Smith offers some down-to-earth advice on investing wisely in the space.

AI dominates the rhetoric of investors, corporate executives, and media, and everyone wants to use, invest in, or acquire the technology. Last year, one in every three dollars invested in the U.S. went into AI startups, and in the current frenetic environment, a seed-stage company could raise $100 million on an AI pitch deck alone.

Goldman Sachs Research predicts that AI investment could reach $100 billion in the U.S. and $200 billion globally by 2025. Microsoft's $10 billion investment in startup OpenAI in January 2023 will be far dwarfed by Sam Altman's latest ambition: to raise $7 trillion to challenge Nvidia in the semiconductor industry.

There has been much talk about the “Fourth Industrial Revolution,” but it is now clear that AI is playing a leading role. Without AI stocks, the S&P 500 would have fallen 2% in 2023. In this new era of unprecedented investment and hype around generative AI and large language models (LLMs), some see the market evolution as eerily reminiscent of the “dot-com bubble” of the late 1990s.

Following the investment patterns of those days, many AI companies are overvalued and overinvested in before they have a viable or profitable product on the market. Even if the AI ​​industry is not yet a bubble, a correction is inevitable, not just for startups training law masters, but for all companies desperately trying to incorporate AI into their business models as if it were the “secret” to success. AI-based business models are not always going to be successful.

Jefferies analyst Brent Thrill summed up the current investor mood: “The hype is there, but the revenues aren't coming,” and “Behind the scenes, the whole industry is trying to work out business models.” [for generative AI]How do we price it? How do we sell it?

But as the initial cash from the 2022 funding rally dries up, the number of AI startups will dwindle over the next year or so, and bankruptcies and consolidations will mean that only companies with truly valuable products will survive. So how do you cut through the hype and invest in companies that are truly profitable?

To avoid succumbing to the same mania that swept through the dot-com era, investors need to be more discerning than ever about identifying companies that are prepared for industry ups and downs, have solid technology fundamentals and business plans that chart a path to sustained growth.

Ultimately, companies looking to disrupt existing markets through the application of AI need to gain market share quickly and protect their ideas as much as possible. Companies that achieve this may be courted by large tech companies, while those that do not will find themselves easily imitated and will quickly lose all advantage if they cannot gain a loyal customer base.

Success is perhaps even more of a challenge for companies looking to create new markets rather than gain share in existing ones. Does the market actually exist? How long will it take to establish it? If you prove a market exists early, how quickly can you scale?

With technology now available to many companies — NVIDIA plans to triple production of its H100 GPUs in 2024 — do new entrants have the infrastructure to run the hardware, develop solutions quickly, and beat the competition? Many forget that it can take up to five years to develop the necessary data center and cloud infrastructure, and the energy to run them. Only companies with access to both the technology and the infrastructure to run it should be seriously considered.

All these considerations are in addition to the usual challenges that startups face, such as securing enough working capital to stay afloat during the development stage.

After all, AI is the future. But capitalizing on this massive opportunity successfully means choosing companies with the technical know-how, market understanding, and access to the critical infrastructure needed to drive this revolution – essentially a “pick and shovel.”

Just as AI-generated fake news has come into the spotlight more and more, investors need to understand that AI labels are not a one-way ticket to profits – but if chosen correctly, the benefits can be surprisingly large.

Rosanne Kincaid Smith is Group COO at Northern Data.

Featured image courtesy of Andrey Suslov

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *