(Bloomberg) — Qodo, a Tel Aviv-based startup that sells artificial intelligence software that tests and detects errors in code, has raised $40 million to expand the business and serve larger customers.
Most read from Bloomberg
The funding round, led by Susa Ventures and Square Peg, brings the total amount raised by the startup to $50 million, the company said in a statement. It declined to make its valuation public.
Software engineers are increasingly turning to AI tools such as OpenAI’s ChatGPT and Microsoft Corp.’s Copilot to generate code, which is then reviewed by humans to filter out errors. Because these tools are trained on public code repositories, there is a risk that they could replicate security vulnerabilities or introduce new ones.
“The more AI helps us write code faster, it also increases the risk surface,” said Itamar Friedman, CEO of Qodo, in an interview.
The possibility that small software bugs could destabilize major infrastructure emerged in July, when a faulty update from Crowdstrike Holdings Inc. millions of computers around the world crashed.
Companies “absolutely cannot risk embracing a high degree of AI autonomy in software development without first having the proper validation and safeguards in place,” Susa Ventures partner Jenna Zerker said in a statement.
While tech companies tout these tools as increasing productivity, the reliance only shifts “the bottleneck” to the tedious work of manually testing and improving code, Friedman said.
Since its inception two years ago, more than 1 million developers have installed Qodo’s tool and several Fortune 100 companies have adopted it. The company, formerly known as CodiumAI, also has offices in the US and the Netherlands. It said it reached $1 million in annual recurring revenue this year after releasing its enterprise software.
Most read from Bloomberg Businessweek
©2024 BloombergLP
