Space and Time has launched an enterprise platform designed to let AI applications verify calculations made on private business data without exposing the underlying records. The practical promise is narrower access for AI agents, alongside a way to check how an answer was produced.
Watch Desk analysis
What happened
Space and Time Enterprise uses the company’s Proof of SQL technology to prove that database queries were computed correctly. Businesses can now run the software on their own servers, without operating a blockchain, buying specialised hardware or relying on an external proving service, according to a report by Cryptonews.net.
The platform already works with Google Cloud’s BigQuery. Space and Time says GPU-accelerated queries over millions of rows can be run and proven in under a second, although performance depends on the query, hardware and dataset. Launch partners named in the report include Maple Finance, Indomobil, Trenez and The Data Foundation.
Why it matters
The offer addresses a real tension in putting AI agents to work with company records: agents may need to use the results of a calculation without receiving unrestricted access to the data behind it. Verifiable query results could give businesses, auditors and AI applications evidence that a calculation was performed correctly, while leaving the underlying records private.
There is an important boundary. Proof that a query was computed correctly does not prove the original data was accurate. Bad records remain bad records, just with a more convincing receipt.
Our read
This is a concrete attempt to make private data more usable in AI workflows without simply handing agents the keys to the database. Running the system on a company’s own servers may also lower the operational hurdle for organisations wary of outsourcing the proving process.
The useful test is whether the system performs reliably on real workloads and fits into customers’ existing data controls. Space and Time’s speed figures and product claims come from the company, as reported by Cryptonews.net; they are not independent performance results.
What to watch
- Whether launch partners describe deployments beyond the platform announcement.
- How query performance varies across datasets and customer hardware.
- How organisations check source-data quality alongside query correctness.
Discussion spark: Would proof that an AI agent calculated an answer correctly be enough to trust it with private business data, or should the agent still be kept away from the underlying records?
Sources and evidence
- Microsoft-Backed Space and Time Launches Platform to Let AI Agents Verify Private Data – Cryptonews.net (9 November 2026, 19:21 UTC)
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.