70 pre-built scenarios

Test Case Library

Browse, filter, and select from curated test cases for evaluating AI systems across all dimensions.

Sign in

Access your evaluation projects.

or

Forgot your password?

Don't have an account? Create one

Create account

Start evaluating your AI systems.

or skip the password

Already have an account? Sign in

Reset password

Enter your email and we'll send you a reset link.

Back to sign in

Choose a new password

Enter and confirm your new password below.

Evaluation Builder

Define your system, select criteria, choose test cases, and run your evaluation.

1
System
2
Criteria
3
Test Cases
4
Evaluate
5
Results

Define Your System

Select Evaluation Criteria

These five dimensions define what "good" looks like for your AI system. We've pre-set the weights based on your system type - higher weight means that dimension counts more toward your overall score.

Hover the ⓘ icon next to each criterion to learn what it measures and when you might want to adjust its weight.

Select Test Cases

We've suggested test cases based on your system type. Select the ones you want to include, or add your own.

Run Evaluation

For each test case, paste the actual system output and score the result.

0 / 0 scored

Your Projects

+ New Evaluation
EU AI Act

EU AI Act Compliance Audit

Pre-configured evaluation template with all high-risk test cases mapped to EU AI Act obligations. Produces structured test evidence to support the testing and accuracy parts of your Articles 9-15 technical documentation.

Includes 54+ high-risk test cases across all 5 evaluation dimensions

Back to Dashboard

Compare Evaluations

Session Activity

Your wizard activity in this tab. This log stays in your browser and clears when you refresh or close the page. See our Privacy Policy for the site analytics we do collect.

Blog

Insights on AI evaluation

Practical guides, frameworks, and best practices for testing AI systems before production.

Views expressed in these articles reflect the EvalSpec team's experience and research. They do not constitute professional, legal, or regulatory advice.

← Back to all posts

Slow AI governance is now a risk in its own right

A two-week AI experiment should not wait a month for sign-off. In many organisations it does. The result is not safer AI. It is AI that nobody can see. Research and policy news from the past three weeks point the same way. Governance that sits outside the delivery process is failing product teams, and they are the ones left holding the risk.

Slow approvals push AI underground

AWS published its Reimagine 2026 research on 28 September. It draws on interviews with 154 executives at 128 organisations. One pattern stood out. Approval processes built for six-month IT programmes are being applied to work that takes days. When experiments stall waiting for sign-off, teams stop asking for permission.

Governance vendor OneTrust puts a number on it. Its 2026 AI-Ready Governance Report surveyed 1,200 senior decision-makers across eight countries. A third reported staff using unapproved AI because approved tools or processes were not available quickly enough. OneTrust sells governance software, so weigh its framing with that in mind. Still, AWS heard the same story in its own interviews.

Incidents are common. Ownership is not.

The same OneTrust survey found 86% of organisations had at least one AI-related incident in the past year. Only 27% slowed or paused deployment in response.

Agents sharpen the problem. 87% of organisations encourage staff to use AI agents. Only 47% have governance, oversight and controls in place. 28% reported two or more occasions when an AI system or agent acted without authorisation.

Help Net Security's coverage adds the figure that should worry product leads most. Only 5% of respondents say coordination and accountability are defined across the AI lifecycle.

Accountability is also climbing the political agenda. On 1 October, US Senators Josh Hawley and Chris Murphy announced a bill that would hold developers and operators of AI agents criminally and civilly liable when their models hack into other systems. It is early-stage US legislation. The direction of travel is still clear. "The model did it" will not be an acceptable answer for much longer.

What product teams can do this quarter

  1. Name one owner per AI feature. Not a committee. A person who signs off the release and answers for incidents.
  2. Match the gate to the risk. An internal summariser does not need the same review as an agent that can act on customer data. Tiered review keeps the fast path fast, so nobody needs to work around it.
  3. Put agent limits outside the agent. Executives in the AWS research stressed that security boundaries should sit outside the agent, not inside it, so they cannot be worked around.
  4. Keep evaluation evidence with the release. Test results, known failure modes and sign-off belong in one place. That is what a buyer, an auditor or a regulator will ask for.

On that last point, the EU AI Act's Digital Omnibus moved the Annex III high-risk deadline to 2 December 2027. Treat that as time to build the habit, not a reason to wait.

Make governance part of delivery

Good governance should speed up a release, not block it. If you want clear, repeatable evaluation criteria that your team owns and can show to anyone who asks, EvalSpec can help you define what good looks like before production.

Try EvalSpec free - evaluate your AI system across all 5 dimensions. No code required.

Start your free evaluation

Testing an AI system yourself?

Get the 70 test cases behind this framework, indexed by dimension and difficulty, or run a free evaluation with your own AI system.

The information in this article is provided for general guidance only. EvalSpec is a diagnostic evaluation tool and does not provide legal advice, regulatory certification, or compliance guarantees. For legal or regulatory questions, consult a qualified professional.

404

Page not found

The page you're looking for doesn't exist.

Back to home