Bernstein just released one of its famous 'blackbook' research reports. This one is about how to invest in AI. Here are the highlights.

  • AI is everywhere, but with it comes a host of new challenges.
  • From costs and adoption to computational efficiency and GPU shortages, investors are paying attention.
  • Bernstein analysts weigh in on how the market for AI is likely to develop.

Brokerage firm Bernstein is famous for its “Blackbook” research tomes that dive deeply into important topics for investors. The latest version came out recently and it’s about artificial intelligence

Since ChatGPT arrived last year, generative AI technology has taken the world — and stock markets — by storm. There’s hype, threats to humanity, massive potential productivity gains, and new online experiences. For investors, that’s hard to ignore. 

“The speed at which Artificial Intelligence (AI) in general and OpenAI’s GPT suite in particular has, in the last six months, taken over seemingly every conversation was the genesis of this Blackbook,” Bernstein analysts wrote recently in an email to clients.

The firm’s analysts identified where, why, and how AI is going to create, destroy, and transfer value in different industries. Then they picked companies in their sectors that could win and lose over the next three to five years as a result of AI. Finally, Bernstein looked for contrarian ideas by asking whether everyone has “simply lost their minds (again)?”

Here are some highlights from the 300+ page report. The full research is available from Bernstein: 

  • Spending on AI-centric systems will reach $154 billion this year.  
  • While there’s a shortage of GPUs for training AI models, the processing power of these chips is improving. “Presumably, the cost curve will slope downward as the scale of LLM adoption rises, but the multifold expansion of the GPU market feels inevitable if LLM usage (whether for search or otherwise) becomes a sustained thing,” Bernstein analysts wrote in the report.
  • If demand holds and services like ChatGPT reach anywhere near Google’s current search volume, there will be sustained demand for GPUs. That’s good for Nvidia, which has a lock on this market at the moment. 
  • AI models will get more efficient, but 4 cents per query multiplied by billions of queries is prohibitively expensive. Only large incumbent companies can afford that and take full advantage of the opportunities. So, we are probably in for big tech dominance, propagated through AI. 
  • The report cited a 2020 CFA Institute study that found only 5% of CFAs had used AI for risk management. The technology’s tendency to “hallucinate,” and companies wielding black box propriety models, can prevent investors from being able to show their work or explain outcomes.
  • AI-powered hedge funds have a history of outperforming the broader hedge fund index, with a 9.8% compounded growth rate versus a 5.8% CAGR overall, which suggests using AI to invest will increase. 
  • Losers in the AI explosion could be data-rich smaller retail brands, think Stitch Fix, that lack AI capabilities and scale, along with big, legacy brands, like Nordstrom, that have the scale but not the technology. “The ones that try to do too much at once and fail to integrate the technology into the core operations of the business” will lose out, the analysts said.

  • AI could help e-commerce companies cut operating expenses through platform and productivity efficiencies, but it’s a double-edged sword with a few downsides and barriers. The direct traffic that the biggest online shopping destinations rely on could be usurped by AI.
  • Online marketplaces could still facilitate the underlying sale, but in losing direct traffic they would risk losing customer loyalty, access to data; and the ability to sell through additional services such as advertising.
  • Online travel agents stand to benefit from AI increasing relevance and accuracy, but the technology is also poised to shake up the sector’s revenue model. Paying to rank high among chatbot recommendations would provide an obvious boost. “Whether OTAs or other tech players win will ultimately depend on who has the best customer data to feed into the AI tool and provide the best offer.” Companies in this sector that have gained market share from 2019 to 2022 have all invested in AI.