Why Most Portcos Are Far From Being ‘AI-Ready’
GPs’ expectations for generative artificial intelligence to drive business growth are high. Do they match portcos’ realities?
A majority of GPs (59 percent) believe that generative artificial intelligence is poised to drive “significant value creation” at their portfolio companies.
It’s a figure that mirrors the hopes of the rest of the economy. According to a Forbes survey, 60 percent of businesses expect AI to increase sales and 59 percent think it will cut costs.
Most private equity portfolio companies, though, are far from AI-ready.
“Not surprisingly, there may be a lot of work to do before the organization’s data and processes are ‘AI-ready’,” explained Scott Kosch, senior partner NextAccess, a management consulting firm that advises many PE firms and their portfolio companies. He noted that for some firms, AI readiness could take upwards of one year or more.
What’s the hold-up? Poor data hygiene.
“When an organization wants to analyze their internal data using AI, we must ask how good their data is. How is their data organized? How do they access their data? Are the AI tools they want to use compliant with the security level they require to protect confidential information?” Kosch explained.
When it comes to compliance, Kosch believes there is actually an overemphasis on the risk posed. “For many AI use cases, like market research and strategic planning, a lot can be accomplished without exposing confidential information to the cloud,” he said.
What organizations should be focused on, in reality, is prepping their data processes now for later implementation.
Kosch added: “How do I secure my data? How do I make sure the data is high quality? That's AI readiness for tomorrow that organizations need to be asking themselves today, or they will soon discover they are behind their competition.”
These improvements have taken much longer, though, than some would like. A 2015 study from Experian found that 84 percent of businesses planned to implement data quality solutions soon. Nearly a decade later, in 2023, The Harvard Business Review found that only 3 percent of companies’ data met basic quality standards.
This work to clean up data at the portco level will likely be painstaking, time consuming, and involve direct collaboration between operating partners and the c suite. Most of AI’s benefits will likely come only in the long term, then.
Despite this, some industry players are still confident that near-term value creation is possible, too.
“Gen AI will likely have tons of potential impact to portfolio companies and their value both near term and longer term,” said Raj Bakhru, co-founder and CEO of BlueFlame AI. “Everyone’s still wrapping their heads around that.”
BlueFlame AI is specialized in bringing AI solutions to alternative asset managers. He noted security detection as an easily implementable solution for GPs and portcos alike.
“If you think about the tooling and the security detection space, a lot of that is large heuristics. A lot of those heuristics require reading and understanding of log files. So large language models actually present a really interesting opportunity to leverage semantic understanding against the log entries to identify anomalous behavior,” he explained.
According to Forbes, cybersecurity is the second-most popular AI application for businesses after customer support, with 51 percent of businesses (second to 56 percent for customer support) either currently using or planning to use AI for it.