• Artificial Intelligence can help law enforcement agencies address COVID-19
• AI needs smart, creative human deployment to be successful
• AI businesses are out there now with rapid deployment models helping law enforcement and government agencies to work their highly stretched resources as effectively as possible

Artificial Intelligence (AI) can help to tackle the major societal issues that have been raised by the COVID-19 pandemic and the potential, local, national and international threats that our governments are facing from rogue operators or the necessary buy-in needed from society as a whole for some of the measures that governments are having to impose during this crisis. AI in and of itself will not make the difference on its own, it is only with the knowledge and creativity of the people who deploy and manage it, and the speed at which it can be implemented can the power of AI be brought to bear.

COVID-19 has the potential to expose some of the shortfalls and limitations of AI. Commercially available AI tools today are forms of Machine Learning, which in essence are sophisticated algorithms for identifying predicted human behaviour by learning from past historical data.  AI has the potential therefore to enable efficiencies and smarter organisational behaviour by detecting patterns that people may have missed.

For any AI system to be effective then, large amounts of rich data sets are needed with the correct segmentations or classifications within them. This is where the first major problem with AI can start to arise – as Machine Learning assumes that conditions today are the same as the conditions that are used within the data sets that it uses to build its propensity models – in other words AI systems implicitly assumes that what has happened historically will happen now and in the future.

These are unprecedented times that we are living in, and with the situation changing weekly – what has worked in the past may not work today. Some of the things we need to try will have never been tried before.

Human insight will bring the most out of AI

It is only with human knowledge and with our ability to draw on our abstract thinking and knowledge from separate settings that we are able to apply these learnings to novel situations.

Therefore, AI will be at its most powerful and useful at any time – and especially during the time of COVID-19 – when it involves the collaboration of humans in several different roles. The data scientists who build the smart Machine Learning systems and bring them to life, will need to work closely with subject matter experts who can help to identify the nature of the problem and some of the issues that may be being thrown up on the front line today – alongside the out-the-box thinkers who help to push boundaries and move us beyond ingrained thinking and habits.

There is a plethora of data on the internet which is open source and utilizing tools which are able to exploit the data and spot trends or given sentiment within a local geography or social group can be very powerful. If deployed correctly and the right insights applied by the data scientists in collaboration with the domain experts, then smarter deployment of scarce and overworked specialist resources can be enacted in very rapid timeframes.

One such SaaS AI business that we have been working with recently was able to deploy its services within a matter of days, and so could bring actionable insights for how a local law enforcement agency was enforcing its fines for breaches of social distancing measures, and the requisite trust it was gaining from the local population.

Therefore through bringing together the technology, with data scientists, subject matter experts and those that challenge pre-existing norms, a better solution and service is available to all.