Open letter to AI Startups and AI researchers

Update 11.2020: It is clear now that the crisis is pushing digitalization needs with an unprecedented demand across all fields and that the shape of the company, the world and economy will look very different from the one before COVID. The way we work is changing, the way we interact is changing, the way we buy and do things are also different and slowly are evolving to a new normal state.
For a company to survive in the new normal, there is an imperative need to foster an organizational culture of rapid adaptation through digitalization into all domains. Also, to create a work environment with less top-down structure of control and with silos of information, towards a team work collaboration, based on trust, with clearer purposeful aims and encouraging the teams in achieving a better emotional quotient within it. The more receptive to these aspects the companies are, the more the better the transition to the new normal will be.

Original text 03.2020: The actual COVID-19 crisis has a lot of lessons to teach us. And we have nothing learned from past pandemics. We’re just falling a downward spiral of human and economic depression. We are living with the hope that this fall will end soon and that we can recover fast. This remains to be seen but besides the many scenarios, we can say for sure that we (as EU to local administrations) were not prepared for the pandemic fight.

My criticism here is simple and only towards the AI field in which investors, CEOs, funding parties, stakeholders, must accept the failure in serving the vital interests, human needs and a failure to set these priorities. In old programming therms, it’s like we are developing a nice GUI, with no good business logic, no solid platform, obsolete middle tier connections. There is no way to advance here until we are not addressing the base layers.

Every single AI researcher should from now on follow a new AI ethic book and focus on what is essential first. The AI research field is so stretched in a multitude of domains and situations while no field excels actually. You can imagine how would have been if all the concentrated AI efforts were to be focused on health-care, fully automated hospitals and personal-care assistants & anti-virus development first. And only after that worked, to try to develop AI based sex-dolls, what a shame to all of us!

I feel myself uneasy now, regretting that I should have pursued what I started in the first place, AI medical informatics with agents that help fight addictions and the build of persuasive personal agents to help for a better healthy life of the individual. With the good AI.

In a hope that this crisis does not deepen and throw us in a new AI stone age we were just escaping, I expect that things will change, and a better evaluation of our AI priorities will be taken into consideration. Maybe a new AI authority will have to take the helm and supervise the further development.

C.Stefan – March 2020.

Big Data Pitfalls

Avoid Simpson’s paradox:
This paradox refers to a phenomena where the association between a pair of variables (X; Y) reverses sign upon conditioning of a third variable, Z regardless of the value taken by Z. If we partition the data into subpopulations, each representing a specic value of the third variable, the phenomena appears as a sign reversal between the associations measured in the disaggregated subpopulations relative to the aggregated data, which describes the population as a whole.

Right ML algorithms usage: use the right approach for machine learning algorithms, find the appropriate algorithm for your specific problems. Ex. If you need a numeric prediction quickly, use decision trees or logistic regression.

Keep in mind the Prisoner’s Dilemma: like in “cigarette manufacturers endorsed the making of laws banning cigarette advertising, understanding that this would reduce ad costs for parties and increase profits across the industry”, so it is with the business strategy and down to big data processing.

Consider Gödel’s Theorem: any system of computation you can construct (numbers theory etc.) that it is true, it cannot be ultimately proved from the rules within that computational construct. The system in a way transcends itself. Thus the way to the strong AI for example.

Keep in mind the exponentially powerful quantum computers of the future. For example build different, resistant cryptographic algorithms against the qubits future powers.