Editor’s note: This article is part of an ongoing series in which Crunchbase News interviews active investors in artificial intelligence. Read previous interviews with General Catalyst, Bessemer Venture Partners, Accel, Insight Partners, Index Ventures, Sequoia Capital, Section 32, M12 and Sapphire Ventures as well as highlights from these stories from 2023.  

AI isn’t a feature. It’s not even a sector, if you ask Dharmesh Thakker.

“I look at AI as a fabric,” said Thakker, a general partner at storied venture firm Battery Ventures.

Thakker recently met with us at his office on Silicon Valley’s Sand Hill Road to talk about the current investment landscape and the 40-year-old firm’s approach to investing in AI.

With AI, “you’re automating human intelligence,” which could be a higher order revolution than cloud or PCs, he said.

Battery has made investments in Arize, Weaviate and Galileo in technologies that assist companies with model management and performance. It is also an investor in Databricks, a more established data products startup that reached $1.5 billion in revenue run rate in July 2023 and acquired MosaicML for $1.3 billion that same month.

Thakker joined Battery in 2015 to build out its infrastructure, data and cloud practice. An engineer by training, he started his career interning at semiconductor companies, then switched to software and was an early employee and co-founder of smaller companies in infrastructure and security that were later acquired.

Prior to Battery, he was the managing director for cloud and big data investments at Intel Capital, the active investment arm of semiconductor company Intel.

“The technology revolutions we’ve seen so far — whether it’s cloud or mobile, or even personal computing or internet — took some inefficient mechanism and tried to provide some technology to make it easier,” he said.

AI goes much further. “The promise of AI is a stake in human judgment and trying to automate some of it so that humans can focus on higher-order tasks that are much more fruitful,” he said.

The impact of AI is also being immediately felt in areas of the economy that have traditionally been further removed from deep tech innovations, he noted. “Increasingly, the boundaries between infrastructure and applications are converging in some sense, because AI is a highly technical space, but the impact is being felt by designers and marketing folks,” Thakker said.

Increased productivity in tech and sales

A historically low interest-rate environment for many years gave the tech industry leeway to become inefficient and less disciplined, Thakker said, but that’s now changing quickly. There’s now pressure on companies to manage costs, he said.

“Just in front of our eyes, we’re seeing companies that are growing 20% to 30%, can now be cash flow break even almost immediately, or within 12 months, by applying generative AI to sales and marketing, R&D, recruiting and other areas of the business,” Thakker said. “And so, we’re seeing this innovation fully impact productivity in the tech sector.”

Thakker notes that it’s estimated more than $500 billion per year is spent on sales and marketing personnel in the software industry, using tactics developed 20 years ago.

But as more companies have seen their growth slow from, say, 50% to 20%, many are asking why they need so many salespeople.

“You can automate half of that by using automated email generation, follow-ups and conversational parts, and engage customers and provide a better experience,” he said. “The ratio of all the overhead to quota-carrying sales reps has dropped dramatically.”

A company generating $100 million in revenue, which was previously spending $50 million on sales and marketing, can now spend $30 million, he said.

Battery portfolio company Gong, founded in 2015 to make sales executives more productive by training and providing conversation intelligence, has improved its technology with generative AI for customers to automate outreach to prospects and follow-up, he said.

Another large cost center, research and development, is materially impacted.

“You can generate output from expensive software engineers making $300,000 to $400,000 a year using Copilot. You don’t need so many software engineers,” Thakker said.

In other sectors such as healthcare, for instance, the impact might be a lot more profound.

Another of the firm’s portfolio companies, Machinify, is used in the antiquated health insurance business to process claims with greater speed and accuracy. According to the company, it has reviewed $200 billion in claims annually for four of the top 10 health insurance companies.

Cloud, semiconductor, AI competition

The so-called “magnificent seven” tech companies — Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia and Tesla — have seen their market values increase $4 trillion to $5 trillion in the space of 14 months, Thakker noted.

“The power struggle between Microsoft and Nvidia, Amazon and Nvidia is about to rise,” he said. Nvidia is building its own cloud service, while Microsoft and Amazon could build their own chips.

In the previous tech generation, Apple built a chip for the iPhone to lessen its reliance on Intel. “It wasn’t the best chip in the world,” Thakker said. “But it was perfect for the iPhone, with low power consumption and graphics, but less computation.”

The dynamics among these leading companies is about to change, even between Microsoft and Amazon, he said. “Microsoft was the enterprise company. Amazon was the one with developers. And Microsoft, by locking up OpenAI, has become closer to developers.”

There’s an opportunity for these $100 billion to trillion-dollar companies to reshuffle the drawing board, Thakker believes.

“Maybe Databricks or Snowflake becomes the third cloud provider to compete with Amazon and Microsoft, while Google’s falling behind.”

“I don’t expect more than 10 to 20 companies to win, but these 10 to 20 companies are going to be worth $500 billion to $1 trillion as opposed to $10 billion to $50 billion. The winners are going to be much larger than we have seen in previous generations.”

This is beyond where startups are competing. However, the software spend is expected to increase as well.

AI leads to a rising wave of software

“My prediction is that the software by itself will expand from $600 billion to more than double or triple that,” in the next five to seven years he said. “So there’s plenty of upside for software companies by automating human labor and delivering more productivity per capita.”

“You could improve GDP by 5% by using technology, you create $5 trillion more value, but only spend $2 trillion more,” Thakker said. “So the software buy could expand dramatically.”

Thakker predicts it’s a rising wave with many companies participating, and not necessarily at the expense of each other. “But ultimately it’s the drug discovery companies and the education companies and the sales and marketing companies that use AI at the application level that will generate that value,” he said.

Related Crunchbase Pro query