An executive’s guide to cognition
Introduction
This executive’s guide to cognition explains why businesses need to harness the power of both human and artificial intelligence. This is particularly important as organisations in their race to become more intelligent are more often than not ignoring the cognitive potential of their people.
Cognition defined
Cognition is defined as the process of making sense of our environment and acting accordingly. Any organism that is unable to do this will not exist for very long.
I am seeing a theme transition in the market from digital business to cognitive business.
Thus cognitive / cognition will increasingly become a marketing term. We already see this in the way that artificial intelligence (AI) is splashed about in respect of product and service offerings. Or how machine learning and AI are used in the same sentence in such a way that the writer clearly doesn’t know the difference.
Focusing first on technology cognition, as this is what cognition vendors are generally referring to. It can be thought of as an attempt to mimic the human brain. The human brain takes its inputs from the body’s senses and interprets what this dataflow might represent, eg. a fast approaching lion. It then decides what action to take based on that perception. AI endeavours to process typically large volumes of data and draw insights through learning, eg. the ability to differentiate between a car registration number and a person wearing a t-shirt with an identical sequence of alphanumerical characters. This might seem trivial to us humans, but like jogging up a flight of stairs, this is highly demanding from a technology perspective. Hence, fifty years in we are still impressed by robots that can for example open a door using the handle (rather than built-in laser gun technology).
In my view, one of the most wasted assets is human cognition.
Our ability to make full use of our brains, which are essentially organic computers that have been programmed continuously for millions of years. The industrial era denatured our cognition, made us all cog workers. We were literally cogs in a (factory) machine. Thus, we had to follow the operations manual and suppress any urges in respect of curiosity and creativity.
Harnessing human cognition results in greater innovation and deeper customer experiences. Humans can read other humans in ways that technology cannot at this point in time. Treating humans as process cogs squanders this cognitive asset.
Buzzwords
It is worth clarifying a few related terms:
Algorithms (aka algos) – In broad terms this is a set of instructions, such as a cooking recipe. It is also the manner in which software is written. Software algorithms can mimic intelligence, as in number plate recognition, or they can be unintelligent, as in sorting a list of names alphabetically. In the former case, the software learns and in the latter the instructions are pre-baked.
Intelligent – Being intelligent means that one can learn and solve problems. If your phone has lots of useful apps, it is smart. If it can configure itself based on your behaviour, it is intelligent. AI is driving the leap from simply smart to intelligent.
Robots – Robots are largely physical devices that primarily offer movement. Examples include the robot arm that paints the car on the conveyor belt, or a carpet cleaning robot. There is an intangible and unmoving class of robots called chatbots that are used typically to mimic call centre operatives in customer service interactions. Robots may be loaded with AI software and so exhibit a degree of intelligence, eg. care home robot. But that is not a given, eg. car plant factory painter robot.
Machine learning (ML) – Can be considered a branch of AI that focuses on learning. It picks up on patterns in the data and draws inferences. It uses statistical techniques to achieve this. ML has given AI a new lease of life in recent years. But it is important to point out that whilst this approach exhibits intelligence, it is not replicating the manner in which the human brain works. In that sense it is a maths parlour trick. Today, you might say we have artificial artificial intelligence (A2I). Until we see the discipline of neuroscience mature, we are unlikely to see any real advances in true AI. It is important to mention this because the majority of discussion in the market today conflates AI and ML.
Why you need to know this
In the industrial era we managed to create a form of synthetic certainty. Thus we could treat the world as predictable. This enabled us to take a factory model approach to our organisations. We needed humans to behave as process cogs. Kaizan and Lean Manufacturing are two approaches that highlight this.
Article: 3 reasons you might be running a factory
The era of synthetic certainty is over.
We now live in increasingly volatile and uncertain times. Therefore. we can no longer predict the future and so the emphasis needs to move away from strategic planning to situational awareness. Business today happens in the here and now. Anything else is fiction.
This new reality requires the introduction of intelligence into our organisational models. Without this intelligence, we cannot make sense of our environment and act accordingly. You can think of this as flying blind in the fog of war.
Innovation is critical to success in this new digital age. The technology is not there yet in being able to do it unaided. That is why we need to use AI to augment the natural cognitive capacity of our people. In effect, turning them into innovative superhumans.
Article: 7 Steps to a thriving workforce
Keep in mind, it is still very early days for chatbots. In many respects it feels like one is dealing with a call centre that is restricted to a diet of barbiturates. But the technology will evolve. However today we need humans to handle the human element (real-time empathy) and make contextual sense of the situation. Personalisation through AI is already well advanced, as we have seen in respect of Facebook and Cambridge Analytica.
Whilst AI draws inferences from large datasets increasingly well. Humans, through millions of years of survival programming, can pick up on very weak signals in small datasets, for example, the snap of a twig on the perimeter of the camp. Humans and AI naturally complement each other.
Again today humans are underperforming in terms of their true cognitive capacity. AI has the power to exponentially unleash the value your people can generate. Keep in mind that AI in its true sense is overrated. That said, human intelligence plus AI is the formula for organisational success in an uncertain world.
Think of cognitive as a new asset class.
Misconceptions
Again, AI is generally overrated, but its time will come when neuroscience matures as a discipline.
AI in the workplace, particularly when embedded in robots, is largely restricted to process work. The AI melts down when faced with ‘exception handling’. But there is no reason why the underlying algorithm cannot learn as it encounters each exception. But even so, case-based reasoning is really a lightweight version of AI. It would be better if the technology didn’t need to get it repeatedly wrong until it eventually got it right. Dental surgery comes to mind.
It has been popular to think of AI / robots versus the humans. My point above is that we should be thinking technology plus humans.
Your reference point might be the film Terminator, where the Skynet computer decides that humans are a problem. In which case, observe your carpet cleaning robot and consider whether it is having thoughts of insurrection. Malevolent AI is of course possible and so regulation is needed. And yes, such software could harness the Internet to hack other systems. But certain systems such as nuclear plants are kept offline for this reason. Again, this will require malevolent human plus malevolent AI.
As mentioned, the terms smart and intelligent are being bandied about. Are we to think, by the definition above, that a smart city is one that simply has lots to offer? Possibly we should be talking about intelligent cities? And possibly we should factor in the humans and their intelligence as we build such cities?
Your five-point data plan
Here are five action points to keep at the forefront of your mind as you develop a cognitive organisation:
- Develop a people cognitive management plan. It is quite likely that you are squandering cognitive capital by having humans doing inhuman process work. Plus, you may have created an operation that stresses your people in some way and thus leeches their cognitive capital, which they spend on managing this stress rather than on creative endeavours.
- Build business models with cognition at the core. It is important to have multiple business models (polymodal business) to offset the risk of your main business being taken out by the digital tsunami. I encourage you to create new business models (think of them as experiments) built with people augmented by technology at the core. Forget the factory model, the game is to build a living sensing organism.
- Keep track of what the big players are doing – Amazon, Apple, IBM, Google and Microsoft. Keep an eye on the OpenAI initiative. Given the real-time nature of doing business on the digital savanna, it would be wise to watch what the surveillance players (most governments), behavioural analytics players (particularly in social media) and moving objects players (drones and cars, eg. Tesla) are doing.
- Be wary of cognising your existing business. Such an upheaval might threaten your primary cashflows and lead to a revolt. If you have a factory business that generates wealth, then keep turning the handle. By all means use AI to make it smarter, faster and cheaper, but best not to tinker with the people, given they were recruited for their cog worker tendencies.
- Be clear on the link between cognitive and data (‘rubbish in rubbish out’). All AI initiatives will serve to accelerate the demise of your organisation if the underlying data sources are inconsistent, or out of date.
- Harness AI to create a great customer experience. I personally believe your people are more important than your customers. Good people will attract good customers and vice versa. That said, customers are critical to organisational survival. Consider how you can use your augmented people to create augmented experiences for your market.
- I would encourage you to think twice about using chatbots at the customer interface. It’s similar to installing children in your call centre (a very industrial era idea). There may be a degree of novelty in this, but in essence you are passing the responsibility of educating your call centre staff to your paying customers. They are unlikely to regard that as an enhanced experience.
In conclusion
Having read this executive’s guide to cognition, hopefully you will think of AI as less a standalone technology, but more as Augmented Intelligence, as in augmenting the intelligence of your people. AI plus humans is the recipe for success in the digital, or should I say, cognitive age.
Article: An executive’s guide to business transformation
Better cognition leads to better decisions and better experiences for your customers. Becoming an intelligent organisation in effect transforms your business model from an inert and rigid factory to a living adaptive organism. One that can sense threats and opportunities as they appear. This is the only model for the increasing volatility and uncertainty we are facing. In the post-industrial age, there is only now, so you need to be positioned to capitalise on it.
