Friar also discussed increasing agentic AI usage in Europe, while acknowledging that the region trails somewhat in adoption.
- Friar said OpenAI is holding back a next-generation model for more safety and alignment work, without a major shift in compute toward safety.
- She pushed back on reports questioning AI returns, citing Cisco engineers saving 1,500 hours a month.
- OpenAI research showed that weekly active users of Codex and ChatGPT Work among EU small and medium-sized businesses have risen 8.1 times since May.
OpenAI Chief Financial Officer Sarah Friar identified cybersecurity as a potential major growth area, and thinks it will be “a big driver of our business.”
In an interview with Bloomberg Television at the International AI Summit in Dublin on Wednesday, Friar described the business as roughly evenly split between consumer and enterprise, with revenue coming from subscriptions and usage-based services.
Discussing growth across OpenAI’s consumer and enterprise businesses, Friar said, “I think that’s going to be a big driver of our business,” referring to cybersecurity. She did not provide a revenue forecast.
The OpenAI (OPEAZZX) executive also pointed to consumer advertising as an opportunity that is just getting started.
Analysts See Growing AI Cybersecurity Opportunity
OpenAI has been expanding its cybersecurity offerings. In August, it broadened its Daybreak Cyber Partner Program, giving approved security providers access to advanced AI models.
In September, CrowdStrike (CRWD) announced an expanded partnership with OpenAI to secure Codex agents and bring OpenAI’s GPT-5.6 Cyber capabilities to its Falcon platform.
Morgan Stanley analyst Meta Marshall estimated in September that AI agents could help more than double the identity-security market to $60 billion over the next two years, citing growing demand to manage access to enterprise systems.
Meanwhile, according to Investing.com, Wedbush said cybersecurity budgets were being redistributed rather than expanded, with spending increasingly concentrated among broader security platforms.
Friar Addresses AI Returns And Model Safety
Asked about reports questioning AI’s return on investment (ROI), Friar pushed back, citing Cisco engineers saving 1,500 hours per month and faster progress on the company’s development roadmap.
Friar also reiterated that OpenAI is currently holding back a next-generation model for additional safety and alignment work. She clarified that this does not represent a major shift in computing resources toward safety, although overall compute spending is increasing for frontier research and to serve growing customer demand.
European Firms Expand Agentic AI Adoption
Friar also discussed increasing agentic AI usage in Europe, while acknowledging that the region trails somewhat in adoption.
OpenAI research released Wednesday showed that Codex and ChatGPT Work accounted for more than half of output tokens among European Union enterprise customers in its sample.
Weekly active users of these tools among EU small and medium-sized businesses increased 8.1 times since May. However, EU-based firms accounted for only 10% of the top 10% of firms globally, ranked by output tokens per active user.
Cybersecurity Stocks In Focus
Growing concerns about AI-related security risks have helped draw investor interest to cybersecurity stocks this year.
Okta (OKTA) led gains in the group, rising around 170% year to date, followed by Fortinet (FTNT), up 130%; CrowdStrike (CRWD), up 114%; Palo Alto Networks (PANW), up 121%; and SentinelOne (S), up 64%. Zscaler (ZS) posted a more modest gain of about 3.5%.
See Also: Fed’s Bowman Sees No ‘Urgent Need’ For Further Rate Action, Wants Time To Assess After September Hike
For updates and corrections, email newsroom[at]stocktwits[dot]com<