Good morning. If AI felt harder to ignore at work this year, the spending figures now explain why. Enterprise spending on AI reached $176.8 billion by September 2026, according to the State of Enterprise AI report Menlo Ventures released on November 3, 2026. A year earlier the figure was $33 billion. That is a rise of 5.4 times in twelve months.
Start your day with the scale of that. This was not a slow climb. Whole categories of office work moved onto AI systems at once – drafting, coding help, customer support, research, document handling and data work, often inside the same company in the same year.
The survey behind the headline
Menlo paired the market total with a survey of nearly 400 US enterprise decision-makers who together oversee more than $4 billion in AI spending. So the report is not only a big number from a distance. It also carries answers from people who sign for the tools and then have to explain the invoices.
Their most telling answer? The surveyed companies spent about 2.5 times what they had budgeted for AI. Read that the right way this morning. Teams set a cautious budget, staff tried the tools on real jobs, usage grew faster than the finance model assumed, and the bill followed the usage. That is adoption running ahead of plans, a pattern that shows up whenever a tool proves itself faster than a planning cycle can react.
Agents are doing most of the talking now
Here is the detail to carry into your day. On OpenRouter, which Menlo uses as one observable channel for tracking usage, agents generated 80 percent of all tokens as of the report – roughly seven times the human-driven usage on that channel. Agent consumption grew 37 times after overtaking human traffic in February 2026.
Be precise about the scope. OpenRouter is one channel, not the whole market, and Menlo presents it as a window, not a census. Inside that window, though, the shift is clear. Most tokens no longer come from a person typing a question and reading an answer. They come from software acting for a person – checking sources, calling tools, comparing options, drafting, verifying and reporting back.
That connects directly to our brief on assistants being graded on real chores. Once software does the asking, charm stops mattering. Accuracy and clean, structured output that the next step can use without a person untangling it – that is what counts. Our item on robots handling a full night shift in a store points the same way, in physical form.
How smart teams should budget after this
A fixed annual line, written once and defended all year, fits flat usage. AI usage is not flat. When a support team finds an assistant handles routine questions well, volume shifts within weeks. When engineers take up a coding tool, seats spread team to team because colleagues can see the results in shared work.
Staged budgeting fits better. Fund proven uses first. Check results against honest measures – tickets resolved, hours saved, work shipped. Then release the next stage. Treat that 2.5 times overshoot as information, not embarrassment. The demand was real. The plan was written before anyone could see how strong it would be.
The one-line takeaway
AI at work is no longer a pilot project with a tidy budget. At $176.8 billion, up 5.4 times in a year, with agents producing most observed tokens, it is core infrastructure now – and it should be planned, measured and funded that way, starting today. If you run a team, the practical move this week is small and clear: pick one AI use you already pay for, name the measure that proves it helps, and check that measure before the next budget round opens.
Also published on:
- News Today World – https://newstodayworld.org/breaking-news/2026/11/04/prince-mario-max-schaumburg-lippe-enterprise-ai-spending-hits-176-8-billion/
- The American Newsroom – https://theamericannewsroom.com/companies-budgeted-for-ai-then-used-far-more-of-it/
- American Pulse Journal – https://americanpulsejournal.com/american-companies-spent-176-8-billion-learning-how-much-they-want-ai/
- American Morning Herald – https://americanmorningherald.com/morning-brief-companies-spent-176-8-billion-on-ai-this-year/
- Billionaire Herald – https://billionaireherald.com/176-8-billion-later-enterprise-ai-outruns-its-own-budgets/
- Revista Minutos – https://revistaminutos.com/el-gasto-empresarial-en-inteligencia-artificial-llega-a-176-800-millones-de-dola/
- Le Miroir Francais – https://lemiroirfrancais.com/les-depenses-d-ia-en-entreprise-atteignent-176-8-milliards-de-dollars/






















Be First to Comment