Serial Alice Proves the Energy of AI, Powered by Intel® Trust Domain Extensions (Intel® TDX) and Intel® Trust Authority (ITA)

AI is a fast-growing share of data-center electricity demand, yet most organizations still report its energy with estimates: model cards, benchmark figures, or a share of a facility's bill. As sensitive AI moves into Confidential Computing, the platform itself becomes protected and verifiable with Intel TDX, and the energy figure deserves the same standard of proof.

Today we are sharing how Serial Alice delivers that proof with Intel® Trust Domain Extensions (Intel® TDX) and Intel® Trust Authority, turning each AI run into an energy certificate that anyone can verify.

AI's energy should be measured, not estimated, and anyone should be able to check it. Intel Confidential Computing and Intel Trust Authority give every certificate we issue an independent foundation.

NVNelson Ribeiro Vicente · Founder and CEO, Serial Alice

Serial Alice in numbers

How we validate: every experiment is pre-registered before measurement, with a fixed workload repeated under identical conditions. Energy is read from the GPU's own counter, recomputable from the raw samples, and each result is published as a signed certificate anyone can verify.

What Serial Alice does, and why it matters

Serial Alice reads energy directly from the hardware (today, the GPU's own energy counter) and issues a signed, auditable certificate for each workload. The certificate binds the measured energy, the run it belongs to, the attested execution environment, and the exact scope of what was measured. It is recorded in a public transparency log, anchored on a public blockchain, and verifiable by anyone on a public page.

For customers, that means:

• Lower energy costs: prove which configuration really uses less energy per unit of work. In our certified H100 tests, configuration changes alone cut GPU energy per token by 30.7% and 61.7%.
• Sustainability claims backed by evidence: numbers an auditor can recompute, supporting CSRD/ESRS energy reporting.
• Fair comparisons: hardware and providers judged on certified energy per workload, not on data sheets.
• Ready for AI agents: as AI moves from answering to acting, enterprises will need tamper-evident records of which agent acted, where, and at what energy cost.

How Intel TDX and Intel Trust Authority help

Intel TDX and Intel Trust Authority don't measure energy themselves; they make the measurement trustworthy.

The GPU's energy samples are sealed into the attested quote for that run, so the figure can't be altered afterwards or moved to another machine, and the token proves it was measured on genuine Intel and NVIDIA hardware.

Intel TDX runs workloads inside a hardware-isolated confidential VM, while Intel Trust Authority acts as the independent appraiser. For each run, the VM generates a fresh TDX quote, including NVIDIA H100 GPU attestation when enabled, and receives an Intel-signed token certifying platform integrity (genuine, non-debug, TCB, and measured boot). This token is embedded in the energy certificate for third-party verification using standard Intel keys. By enforcing strict appraisal policies, Serial Alice verified reference image matches, boot-configuration changes were consistently flagged, and token issuance was blocked on failure.

Attesting the meter, not just the machine

An energy figure is only as good as the software that measured it. In a pilot on our live service, Serial Alice issues certificates in which one Intel Trust Authority token covers three things at once: the Intel TDX platform, the NVIDIA H100, and the IMA measurements of the energy collector's files, compared with an authorized, signed release.

Before a certificate is issued, ten checks must pass, from the authorized collector release and every measured file, to Intel's replay of the measurement log against the TPM, to the binding of fresh evidence to this run's samples. An auditor can see that the authorized collector release was loaded during that boot. The next step, which we are building with Intel, is to bind each certificate to the exact code that produced it in that run.

Use cases

1. Certified savings under confidential computing. On a Google Cloud H100 with confidential computing inside Intel TDX, 12 pre-registered runs showed 61.7% lower GPU energy per output token for the evaluated attention configuration, with no errors and no quality loss, each run backed by its own Intel token.
2. Certified GPU optimization. A single clock setting on an H100 cut energy per token by 30.7% in a certified outcome. Optimization providers can show customers a saving that is audited, not asserted.
3. Trusted measurement for GPU clouds. Providers can show that the energy on an invoice or report was measured on an attested platform, with the authorized collector release loaded.
4. Accountable AI agents. The platform also supports certificates for agent actions: each action signed, chained to the agent's previous action, and bindable to Intel TDX evidence, with the energy of the work attached.

Built for auditors

• Recomputable: energy is recalculated from the raw samples and must match exactly.
• Independently attested: Intel's token is verified with Intel's published keys.
• Tamper-evident: change one sample, one signature character or the energy figure, and verification fails.
• Portable and permanent: an offline bundle for auditors, a public log and an on-chain anchor.
• Future-proof: post-quantum ML-DSA alongside Ed25519 on supported hybrid certificates.

Try it: open api.serialalice.pt/v2/verify to verify sa-04d80001986249969970fb0db3aec78f, one of the 12 runs.

What's next

Intel's direction for confidential computing (TDX, trusted I/O with accelerators, post-quantum cryptography, enhanced attestation and agentic AI) sets the path for verifiable energy:

• one certificate covering CPU, memory and GPU energy, measured by an authorized collector on an up-to-date Intel platform;
• agent actions linked to who authorized them, and to the energy they consumed.