41 Stats That Tell the Story of AI Right Now

AI Is Now Mainstream In The Workplace
- 52% of US workers now use AI on the job according to a mid-May Gallup poll. Nathaniel Whittemore treats this as the new baseline for workplace behavior, shifting focus from adoption to effective use.
- This broad statistic signals ubiquity across roles, not just frontier teams, and reframes questions toward how well organizations leverage AI rather than whether they use it. (Time 0:02:30)
Value Perceived But ROI Is Still Nascent
- Companies often see value from AI in pockets but struggle to translate that into measurable organizational ROI. Examples: Domino Data Lab, PwC, and KPMG show high reported value but low established ROI.
- KPMG found only 7% of global leaders reported established ROI, while 76% said AI delivered meaningful business value, highlighting a measurement and integration gap. (Time 0:02:54)
Token Costs Are Triggering A Cost-Conscious Era
- Token costs are forcing C-suite leaders to rethink AI plans, with the EY AI Pulse showing 98% of leaders concerned about token expense. Many firms still don’t meter usage.
- Only 64% actually meter token usage, creating risk when agentic workflows drive large, unpredictable consumption. (Time 0:04:00)
Huge Spending Gap Between Average And Top AI Users
- Ramp card data on 70,000+ businesses shows massive internal spending disparities: median AI spend was $11.38 per employee per month versus $7,500 per month for the top 1%.
- Ramp data also showed Anthropic slightly overtook OpenAI among their customers in July, reflecting a concentrated two-company race for adoption. (Time 0:04:53)
Employee Resistance To Agents Is Rising Rapidly
- KPMG’s quarterly pulse found employee resistance to AI agents jumped from 5% to 20% in one quarter, suggesting growing pushback as agents become more capable.
- Whittemore links this rise to the increased perceived threat of agentic systems and flags the result as worth monitoring for adoption dynamics. (Time 0:06:21)
Agents Are Being Given Multihour To Full Day Tasks
- OpenAI found over 25% of Codex users delegate tasks estimated to take more than eight hours of human work to agents, indicating real delegation of substantial tasks.
- Whittemore uses this to argue agent capabilities are expanding the types and size of work being handed off to AI. (Time 0:07:02)
Stigma And Policy Drive Secretive AI Use
- Workers face social and policy friction when using AI: Atlassian found disclosure leads to being judged ‘10 times lazier’ and PagerDuty reported 66% used tools they believed violated company policy.
- Whittemore highlights these stigma and policy gaps as drivers of secretive AI use and enterprise shadow IT. (Time 0:07:27)
Bot Sitting Is Becoming Core Work For Many Roles
- BCG’s June study found 47% of workers report spending more time managing and supervising AI than doing ‘actual work,’ framing agent oversight as a new responsibilities.
- Whittemore argues that supervising agents may itself become the core job for many roles rather than a temporary overhead. (Time 0:09:25)
Most Kids Use AI But Parents Rarely Discuss Safety
- Common Sense Media reports 86% of kids ages 9 to 17 use AI, yet over 40% say no parent has discussed AI safety with them. Whittemore calls this a parental engagement failure to avoid repeating social media mistakes.
- He urges parents to learn and discuss AI with children now to mitigate downstream social and mental-health harms. (Time 0:20:10)