ETHICS / ACCOUNTABILITY

Write the rules before the pressure arrives.

Sevenbitlabs has not started a research project yet. That makes now the right time to say what the work should never compromise and to be clear about the safeguards that still need to be built.

01 / Where we are

A commitment is not the same as a working system.

Still to be built

Sevenbitlabs does not yet have an independent Research Ethics Board.

Before research involving people, sensitive data, health, or other serious risk begins, we need qualified independent reviewers, clear procedures, named responsibility, and a safe way to raise concerns. Until those exist, this page is a promise about what we intend to build, not proof that it already exists.

02 / Commitments

The minimum we should expect from our own work.

  1. 01

    Respect people

    Participation should be informed and voluntary, and no one should face avoidable harm for the sake of a result.

  2. 02

    Make the benefit real

    People affected by the work should understand the question, the risks, and what, if anything, they stand to gain.

  3. 03

    Take less data

    Collect only what the work truly needs, protect it properly, and say who can reuse it and for how long.

  4. 04

    Count the wider cost

    Look at the energy, materials, environmental effects, and community consequences of the whole project.

  5. 05

    Keep humans responsible for AI

    Test for bias and failure, explain important decisions, plan for misuse, and give people a way to challenge outcomes.

  6. 06

    Name competing interests

    Disclose who is funding the work and keep conflicted people from reviewing their own decisions.

03 / A future review process

How a project should earn the right to proceed.

  1. 01

    Explain it

    State the question, why it matters, the evidence so far, the risks, the hoped-for benefit, the funding, and any conflicts.

  2. 02

    Challenge it

    Ask qualified people to test the science, feasibility, safeguards, data choices, and possible effects on communities.

  3. 03

    Watch it

    Track consent, data handling, unexpected harm, and whether the balance of risk and benefit has changed.

  4. 04

    Tell the whole story

    Publish the method, funding, conflicts, limits, failures, and results without making the evidence sound stronger than it is.

04 / Raise a concern

If something feels wrong, there should be somewhere to say so.

We do not yet have an independent reporting channel. For now, use the contact form and choose “Ethics or research concern.” We will replace this with an independent route as the lab grows.