Not too long ago, artificial intelligence was more or less a science project in which data scientists and developers fooled around with one-off small or test projects. Their bosses, CIOs or IT directors, had little clue what was being developed.
Now, it looks like the stakes on AI are being raised to atmospheric levels – and business leaders need to take charge. At least 50% of CEOs recently participating in a Boston Consulting Group survey of 2,360 CEOs say their jobs are the line if their AI efforts flop. That’s quite a change. In addition, 39% of CEOs say they are ready to lead major AI efforts, compared to 38% of CIOs/CTOs. and 25% of tech executives.
Why are CEOs worrying so much about what recently were science projects within the deep recesses of their organizations? Because they recognize that the days of AI’s one-offness are over – it’s now a technology sweeping through and disrupting the entire organization. And it requires some very substantial investments. Currently, on average, 1,7% of revenues is dedicated to AI investment.
Agentic AI is where the action is these days, CEOs overwhelmingly agree. Nine in ten surveyed believe AI agents will enable their organizations to report measurable ROI in 2026. More than 30% of their organization’s AI investments this year will go to agentic AI.
It’s getting personal as well. CEOs now spend eight hours a week spent expanding their expertise on AI, the survey finds.
A separate report out of BCG urges adoption of the “10/20/70” principle to achieve business success with AI: Devote 10% of your resources to algorithms, and 20% to technology and data. The remaining 70% of your resources to people and processes. The problem is many organizations get this backwards.
It takes a redesign of end-to-end processes to help AI succeed. This redesign takes leadership-level executives to oversee change management and an openness to change, the BCG team of authors, led by Eric Jesse, state. They must decide whichever to use decentralized agents or adopt a centralized infrastructure. They must decide whether to invest in pre-built commercial solutions or have internal staff developing and deploying AI. Either way, money must be directed accordingly, with the possibility of making the wrong bet.
On its surface, the benefits of AI agents may make some bets budget-worthy, BCG believes. “Deploying AI agents will execute purely transactional work—rule-based, high volume, low variance—on their own. Other agents will provide an unprecedented level of support to human teams, who will still manage complex, high-stakes relationships.”
Jesse and his co-authors predict a far-reaching impact from agentic AI, with “significantly fewer frontline employees and a corresponding reduction in the management layers.”
As a result, “managers will have smaller teams focused on higher-level, higher-value tasks where humans still excel, augmented by AI tools they know how to use.”
The BCG team offers the following advice:
- Make AI a key priority: Be a disruptor, not the disrupted.
- Deepen AI literacy: Everyone needs to help lead an AI effort.
- Invest decisively: And be sure to fund end-to-end business functions.
- Upskill the workforce: Productivity, creativity, and judgment are essential skills for the AI era.
- Measure progress: Or potential lack thereof.
