Concern 387 · Working draft · editorial review in progress
Bias bounties and audits
Bias audits and bounties are ways to look for unfair AI behaviour, but their value depends on what is tested and what happens afterward.
Understanding the concern
An audit examines a system using a defined method. A bounty invites people to report problems, sometimes for a reward. These approaches can uncover issues a developer missed, especially when people with different experiences participate. Twitter's 2021 image-cropping challenge is an example of a public bias bounty.
Neither approach guarantees that every harm has been found. A narrow test, limited access or poor incentives can leave important questions untouched. It is like inspecting a building: knowing which rooms were checked matters. Useful results explain the scope, limitations, findings and corrective action, rather than simply awarding a reassuring label.
An example
A tool passes an audit that checks one demographic category but never tests disability access or combined disadvantages.
Illustrative unless identified as a documented case.
Words in plain English
- AI
- Software that learns patterns from examples or uses rules to perform tasks such as recognizing images or generating text. It can make mistakes and does not understand every situation as a person would.
- Reward
- A signal used in some training methods to favor certain behavior. A high reward score does not necessarily mean a good result for people.
Evidence and its limits
The bounty account is first-party reporting. Assurance guidance provides broader context; neither makes an audit a universal certificate of fairness.
Twitter Engineering · First-party technical account
Sharing Learnings from the First Algorithmic Bias Bounty Challenge ↗ (opens in a new tab)2021 public image-cropping challenge; company account, not a guarantee that bounties find all harms.
UK Department for Science, Innovation and Technology · September 2025 · Official policy roadmap
Trusted third-party AI assurance roadmap ↗ (opens in a new tab)Proposes professional and market-quality development; a roadmap is not evidence that every service already meets a common standard.
Broader frameworks and background
These resources provide context. They do not establish every claim in this concern.
A question to discuss
What was outside the review, who could participate, and which findings led to a real change?
Keep track of different views, what needs checking, and any next step in your own discussion notes.
Open discussion notes →Principles for the conversation
Before you finish · reflect together
Virtues and principles to consider
Virtues are qualities we practice in how we treat people and make choices. They are not scores for judging someone. Use these ideas to explore what a thoughtful response could look like.
Justice
Justice means fair treatment, fair procedures, and attention to who receives benefits and burdens. It asks whether people with less power can be heard and seek repair.
Look at outcomes across different groups and individual cases. Provide a usable way to challenge decisions and correct harm. Equal wording in a policy does not guarantee equal treatment in practice.
Ask together: Who benefits, who bears the cost, and who can challenge the decision?
Dignity
Dignity means recognizing each person's worth regardless of usefulness, status, or a system's score. People deserve respect and a voice in matters that affect them.
Explain important decisions, avoid humiliating treatment, and keep room for a person's story. Do not treat a prediction as a complete account of who someone is.
Ask together: Does this use treat each person as someone whose experience matters?
Listening
Listening means trying to understand someone's experience before preparing an answer. People who use a service, work with it, or are judged by it may notice problems its designers miss.
Invite people to describe what happened in their own words. Repeat back what you understood and let them correct it. A quiet person or a complaint that is hard to measure may still reveal something important.
Ask together: Whose experience have we not heard yet?
Editorial ratings
These are editorial judgments from the register, not measured probabilities or predictions about your situation.
About the register and scoring →More ways to explore
Search for more writing
For more specific or recent material, open a search below or ask for a web summary. AI-generated summaries can contain mistakes; open the original articles to check them.