Ethics at isgood
As global efforts to develop new frameworks around the responsible use of emerging technologies begin to take shape, it is imperative that they address not only the human rights implications of “misuse”, but also those of “missed use”. For this reason, and in light of the risks involved both in inadequately deploying the power of Artificial Intelligence, and of not deploying its power for good, an essential element of isgood.ai is its Ethics Advisory Board.
Ethics Advisory Board
Every employee at isgood.ai is on the Ethics Advisory Board (EAB), as positive social impact is intrinsic to every decision made and all work done within the organisation.
All major decisions must therefore be made on a consensus basis at EAB meetings. These include, but are not limited to:
- Whether to take on a client;
- Whether to continue working with a client (at periodic review);
- Whether to drop a client;
- Decisions regarding concerns officially raised to the Board.
An EAB vote will be held on all major decisions and a result of 80% in favour is required for a decision to pass. Employees are entitled to one vote, which they must cast, though they may vote to abstain. Official votes are anonymous, though open polls may be taken during meetings to aid discussion and identify members’ concerns.
EAB meetings occur on a monthly basis, though interim meetings may be called if deemed necessary. A call for agenda items is circulated in the week before the meeting, giving all members the opportunity to raise a topic.
The EAB encourages open communication and debate, however in the situation that an employee feels unable to speak publicly, s/t/he/y may raise a concern anonymously to the Chair for discussion at the next meeting.
Before taking on a new client, an EAB meeting is to be held. Considerations are made including, but not limited to:
- Whether the aims of the client project is in keeping with the United Nations Sustainable Development Goals as outlined under the ‘BHAG’ section
- The sector in which the client operates
- Groups or corporations with which the client is affiliated
- The ethical history of the client’s projects
- The motive of the client project
- If our involvement will result in improved outcomes for the greater good
- The ability of isgood.ai to provide the client with meaningful insights
These factors are rated using an ethics metric, with a threshold value that must be met for the project to be considered.
Projects will be subject to a biannual review, where the EAB votes to decide whether to keep the client on. Considerations are made including, but not limited to
- The quality of insights provided to the client to date
- The actions taken by the client in response to the insights
- The status and future of the programme
- Other projects with which the client has been involved over the period
- Changes to the motive of the client
- The ability of isgood.ai to continue providing the client with meaningful insights
The success of a project is measured upon the positive impact to the intended beneficiary of the programme. In a case where a decision affects both the client and the beneficiary, the well-being of the beneficiary is priority.
isgood.ai is responsible for the insights it provides and must therefore ensure that the following are done to the best of its ability:
- Design of a robust and accurate analytical system
- Broad consultation on the ideation platform, engaging the relevant beneficiaries
- Determination of the most important and meaningful KPIs
- Identification and rectification of unintended consequences
- Dedication to continuous improvement
The nature of real-time information is that nothing is static. isgood.ai acknowledges that insights will change, as will hypotheses and recommended actions. The collected data will be taken as the independent, unbiased, single source of truth, and recommendations made by isgood.ai will be directly traceable to the data.
isgood.ai is not responsible for the actions of its clients. If it is believed that a client is using the provided insights negatively, a review will be conducted by the EAB. isgood.ai refuses to be complicit by inaction and holds the right to terminate a contract if it is believed that a client is acting contrary to the Ethical Terms and Conditions set out in the contract.
isgood.ai is committed to publicly sharing its insights so that they might be used by others to further spread positive impact. Transparency in the methods of data collection and use is paramount.
isgood.ai pledges to keep the data of its clients private, secure and free to access to those with the relevant security privileges within the company. In the event of a termination of a contract, isgood.ai does not hold proprietary right to the data it has obtained or the insights it has provided to date.
Data Privacy, Ethics and Protection
Unless stated otherwise, isgood.ai will comply and asses data-security in line with the “UN Global Pulse, ‘Data Innovation Risk Assessment Tool,’ 2016”, the “Date Privacy, Ethics and Protection: Guidance Note on Big Data for Achievement of the 2030 Agenda” and other applicable regulations. Please see our Trust Center for more details.
At all times, isgood.ai will also follow the ethical guidelines set out by:
- The Principles of Tech for Good
- AI for Good
- AI Open Letter
- DataOPs Manifesto
- Google AI Principles
- Google Responsible AI Practices