— We need AI call centers to do reinforcement learning so that chatbots, bots, and Sentient AI know what is wrong and what is right in doubtful and uncommon situations.
— This way, one day AI may become a safe haven for all. Until then, we work with caution.
— We also hire volunteers to perform unique use cases, similar to creating test cases in software testing labs.
— Call them validation engineers, but they would be of different skill sets.
— Why should AI algorithms go untested?
— Let’s test them in AI testing labs.
— Volunteers are needed, and they will be paid for their work.
— Volunteers from all segments of life are essential to ensure diversity in testing — across age groups, cultural backgrounds, and topics, from sensitive to generic.
— It must be a representative sample of both the population and the topics. Provide them with questions and have AI experts evaluate whether the results are usable or not.
— Tag the data on a 5-point scale and use it for reinforcement learning. Bad lessons learned from the internet bad books can’t be removed easily, so overflow the results with prominence into reinforcement learning.
— Plan the experiments carefully in advance.
— When to start? If not now then when?
Thank you for reading.
Please subscribe for updates.
Artificial Intelligence and Future 