Exclusive: CollectivIQ targets AI costs with control platform
Boston-based startup CollectivIQ Inc. is targeting runaway artificial intelligence costs with a platform intended to give businesses control over how models are used. The company describes its product as an "AI consensus platform" that allows administrators to assign employees access to different classes of AI models according to their roles, departments, business requirements and budgets.
Key Takeaways
- Admins can also impose organization-wide spending restrictions and daily cost and token limits for individual workers.
The features address a growing problem for companies that are moving generative AI from small experiments into broader production use.
- CollectivIQ said its new controls expose usage at both the employee and company levels and allow administrators to restrict background and agentic work.
The company takes a different approach from enterprise AI services built around a single model provider.
- CollectivIQ instead accesses models through their application programming interfaces, paying for the tokens consumed rather than a full seat license for each worker.
Its software then routes requests according to their difficulty.
- Companies can customize access for particular employees.
The approach is also meant to address an apparent contradiction in CollectivIQ's design: Consulting multiple models could consume more tokens than asking only one.
- Users can see responses from individual models and make their own judgments.
Stats & Key Facts
- #But providing every employee with a conventional enterprise AI license would have cost about $700,000 annually, he added.
SiliconANGLE UPDATED 09:00 EDT / JULY 27 2026 AI Exclusive: CollectivIQ targets AI costs with control platform by Paul Gillin Boston-based startup CollectivIQ Inc. is targeting runaway artificial intelligence costs with a platform intended to give businesses control over how models are used. The company describes its product as an "AI consensus platform" that allows administrators to assign employees access to different classes of AI models according to their roles, departments, business requirements and budgets.
Admins can also impose organization-wide spending restrictions and daily cost and token limits for individual workers. The features address a growing problem for companies that are moving generative AI from small experiments into broader production use. Reasoning models, background processes and autonomous agents can consume tokens without employees or managers having a clear view of the resulting expense.
Recent high-profile examples include Uber Technologies Inc. burning through its entire annual AI coding budget in just four months and Swan AI Inc. running up a $113,000 monthly bill for a four-person team.
For more details please read the original article at SiliconANGLE AI.
Continue Learning
Comments
Sign in to join the conversation