Developing
No sample size, period or geography is given, and Akamai sells browser-level AI monitoring products.
In 60 seconds
- 47.11% of enterprise AI conversations run on personal identities and 52.89% on corporate accounts.
- 16.31% of AI browser extensions have known CVEs versus 10.8% of all extensions; 18.19% request cookie access versus 6.67%.
- The figures match LayerX's State of AI Usage Report 2026, covered by PPC Land in May.
- The report discloses no sample size, user count, collection period or geography.
Impact Layer
- What changed
- A vendor is re-promoting numbers on shadow AI use in companies.
- Why it matters
- Marketing teams often use personal accounts for AI tools, which can send company data outside company controls.
- Who is affected
- Security, IT, legal and marketing leaders setting AI use policy.
- Martech stack area
- AI, Data
- Stack impact
- Find out which AI tools staff use, through which accounts, and set a policy that offers an approved route.
- Business impact
- Low PPC Land notes the report has chart and text inconsistencies (a 14.4% figure on a different base; median conversations per user 2 or 12) and the claim that corporate-email data on personal accounts feeds public model training has no supporting data.
- Related technologies
- What to watch
- Independent measurements of personal-account AI use.
On October 21, 2026 Akamai emailed a promotion for its Enterprise AI Usage Risk Report 2026, a 27-page State of the Internet edition dated August 2026, according to PPC Land. The data comes from LayerX, a browser security firm Akamai acquired, and matches LayerX’s State of AI Usage Report 2026, covered in May.
Key numbers (as attributed by Akamai)
- 47.11% of enterprise AI conversations run on personal identities; 52.89% use corporate accounts.
- 14.4% of corporate-identity conversations use personal freemium licences (the chart’s base differs from the body text).
- 16.31% of AI extensions have known CVEs, versus 10.8% of all extensions.
- 18.19% of AI extensions request cookie access, versus 6.67% of all extensions.
Gaps
- No methodology: no sample size, user or organisation count, collection period or geography.
- LayerX’s sensitive-data measurements from the May original are absent.
- The chart labels median conversations per user as 2 while the text says 12.
- The claim that corporate-email data on personal accounts feeds public model training has no supporting data.
Our read
Akamai sells browser-level AI monitoring, and its advice points that way. The practical point, finding out who uses what, stands without it.
Sources
- PPC Land: 47% of enterprise AI chats run on personal accounts, Akamai report says (opens in a new tab)
Source for the report's figures, authorship and gaps.