Salesforce is trying to catch up in an AI arms race

Salesforce's Einstein Copilot is unique in its focus on enterprise customers, and the generative AI market is still growing — a lot

We may earn a commission from links on this page.
A sign is posted at Salesforce headquarters on February 28, 2024 in San Francisco, California.
A sign is posted at Salesforce headquarters on February 28, 2024 in San Francisco, California.
Photo: Justin Sullivan (Getty Images)

If the first outlines of today’s generative AI boom began taking shape roughly a decade ago — say, alongside the founding days of OpenAI — Salesforce was early to the scene. The company launched its first proprietary AI tool in 2016, well ahead of our current conceptions of generative artificial intelligence. But it’s since fallen behind.

The cloud-based software company that rivals the likes of Microsoft and Oracle just recently launched its first generative AI assistant, Einstein Copilot. It came well after OpenAI’s ChatGPT, Google’s Gemini (formerly Bard), Microsofts Copilot, and Anthropic’s Claude debuted. But luckily, the tool first announced by Salesforce last fall has an edge in one department: it’s marketed itself as a tool solely for companies, not consumers — a more specific corner of the market where it’s better positioned to thrive.

Advertisement

Arion Research analyst Michael Fauscetter told MarketWatch Monday that “Salesforce has a lead” in the market for enterprise generative AI tools. Einstein Copilot can complete a range of tasks, from initiating returns for customers to recommending them new products.

Advertisement

Even in the broader AI field, there’s still (arguably a lot of) room for competitors like Salesforce to make an entrance. According to a recent analysis from Bloomberg, the market for generative AI is set to reach $1.3 trillion by 2032.

Advertisement

Not to mention, existing AI tools remain flawed. Microsoft Copilot and Google Gemini, for example, are still making major mistakes. Google had to shut down Gemini briefly after it produced historically inaccurate images. Microsoft Copilot and Open AI have faced heat for generating violent and copyright-violating images. And the biggest of big tech companies are scrambling to fend off lawsuits over how they train their AI models. Last week, French regulators fined Google for training Gemini on content from French publishers without their consent. And The New York Times sued OpenAI and Microsoft last December over a similar issue.