Ann: Appen 2020 Full Year results, page-232

The data will still need to be analysed by humans. From a regulatory and general standpoint, we are not at a point where we trust AI/ML to make decisions for us. One thing to be cautious of is these AI/ML circuitboards are likely to subject to heavy testing and regulations. It starts to become very interesting when AI/ML is allowed to make decisions for humans since they evolve and start to create their own inputs as a derivative of their outputs, which generates unintended outputs. Unfortunately, these circuitboards don't know when to stop learning.

While these technologies can help with data annotation it is no different to using other forms of AI or ML to help automate some of these tasks which Appen is currently looking at.

Neuromorphic chips are the future of AI but they are also the next step towards artificial super intelligence, so will likely receive a lot of scrutiny and regulatory pressure.