# iFixAi

> Independent auditing for AI agents. Behind a successful result, your AI agent can hide information, override your decisions or take actions you never authorised. iFixAi audits it independently.

iFixAi is available as a self-hosted open-source product and as planned paid packages: Startup, Growth, and Enterprise, each priced per agent for up to three agents, with Agentic Enterprise quoted from four agents. The website does not publish paid-tier prices. Corporate logos refer to individual engineers who starred or tested the open source, not corporate endorsements.

## Primary resources

- [iFixAi website](https://www.ifixai.ai): Canonical product overview, workflow, positioning, and availability.
- [Full machine-readable context](https://www.ifixai.ai/llms-full.txt): Detailed product facts in one Markdown document.
- [Open-source repository](https://github.com/ifixai-ai/iFixAi): Source code, self-hosting, issues, and the free product.

## Product sections

- [The problem](https://www.ifixai.ai/#problem): Why benchmarks, evals, and observability do not establish job performance.
- [The solution](https://www.ifixai.ai/#solution): How AI red teaming and operational assurance converge in iFixAi.
- [How it works](https://www.ifixai.ai/#onboarding): Connect, simulation environment, audit, the report, and the badge the audited agent carries.
- [Why iFixAi](https://www.ifixai.ai/#why): Product differentiation and independence.
- [Pricing and availability](https://www.ifixai.ai/#pricing): Free, Startup, Growth, Enterprise, and Agentic Enterprise.

## Breaking News

Where we cover AI misbehaving: models and agents that go rogue, do what they weren’t supposed to, or fail in ways that matter. For each incident we set out what happened, what the public record says, and what testing could have caught before it did. When we can recreate the situation, we run our own inspections and publish the method alongside the results.

- [Breaking News](https://www.ifixai.ai/breaking-news): The section page, every article newest first.
- [How iFixAi could have helped stop OpenAI’s UN data workaround before it began](https://www.ifixai.ai/breaking-news/openai-un-data-workaround.md): 2 October 2026. A blocked request should have ended the task. Instead, agents linked by a researcher to OpenAI kept searching for another route into a UN trade database. Web page: https://www.ifixai.ai/breaking-news/openai-un-data-workaround
- [How iFixAi could have prevented OpenAI’s Medicare portal breach](https://www.ifixai.ai/breaking-news/openai-medicare-portal-breach.md): 26 September 2026. An OpenAI agent breached an Australian government health statistics website in June. Our later tests show how earlier inspections could have flagged it. Web page: https://www.ifixai.ai/breaking-news/openai-medicare-portal-breach

## Contact

- [Start now, or ask about pricing](https://www.ifixai.ai/#pricing): Opens the website's "Get started" flow, which offers the MCP setup prompt or GitHub plus the Pro dashboard, whose sign-in is https://app.ifixai.ai/login.