Update- The Promise of AI on Everyday life.

Update- The Promise of AI on Everyday life.

I wanted to do an update on this article I wrote last year to see that has changed regarding the status of #AI as it relates to the products we use everyday and what I might see at #CES2023 this year. 

We all believe products should be smarter, and often disappointed when we have to enter our email address twice, change apps to find our last appointment, are forced to find the traffic lights or crosswalks over and over in the Captcha security box, or reset all the positions in our car after lending it to a friend. AI, also referred to as Machine Learning, is just that, machines that learn. Applying techniques that allow Machines to learn (think of a machine as any product or service we might use) should help these machines get better and smarter over time, helping create a more natural and anticipatory usage experience. If you want to learn more about Machine Learning, go here

What is the biggest change in the past year? ChatGPT: I’m sure by now you’ve all tried it. If not, head over to Openai.com, “Try ChatGPT”, create an account and you’re on your way. It can answer questions, write letters, or even write a screenplay, or college essay (lots of discussion about this). There are many articles about ChatGPT that I won’t try to duplicate here, but if you want s simple guide, this one is as good as any.

Context will make products smarter and more personal: Products don’t “know” anything about you or what you like- your toaster doesn’t know how you like your toast, your Garmin bike computer doesn’t know it’s raining, and Alexa doesn’t know you’re frustrated. Context will make products feel like intelligent partners, rather than stubborn mules- how you hope and expect them to behave a certain way, and they just don’t. 

Natural interfaces will replace rigid ones: People want products to be more like people- to understand them and as such respond to more human like gestures. When a friend asks in what direction is the nearest bathroom, you just point. BMW uses gesture control and Synapse Product development has prototyped a more natural interface for cooking. While voice has its place in scenarios like driving, chatbots, Digital assistants (ie Alexa) and other interfaces where voice solves an obvious interface problem, it’s not the best interface for all interface occasions such as where privacy and social norms prevent the wide scale use of voice. The quest for the perfect interface will continue and we must acknowledge that one size fits all does not apply here.

Predictive algorithms will keep us healthier: Our sleep patterns, how we exercise, what we eat, even how we walk and talk are all influenced by, and predictors of health. This data can be used to suggest behaviors and actions to improve our quality of life. Whoop for example, a company founded by two Harvard grads, has a wearable device that uses heart rate variability to monitor your performance and suggest when to rest or, recover, or get more sleep. I’ve written more about data in healthcare here.

In the areas of human performance, using data to predict future performance can help athletes train today, for impact tomorrow. Smart devices and systems that can read physiological data and use it to predict human performance is a hot and booming field. Companies like Whoop, are developing systems and algorithms to use various biomarkers such as heart rate, heart rate variability, skin temp, breathing rate, and SPO2 in order to give athletes a view into their bodies like never before- ie your Digital Twin. If interested in Digital Twins for consumer products, there is an article here to learn more.

AI will provide MORE opportunity for surprise and delight: Not many in industry address this topic as much as bias or ethics in AI, but the challenges or an overly curated life are something to be considered. While many fear AI will product an overly curated life (ie Amazon- “since you bought that, you might like this”, and Spotify suggests songs like the ones you’ve listened to most), it doesn’t have to be this way. AI can expose you to new experiences you didn’t even know existed if developers can create algorithms not just based on history or preference, but by encouraging change and surprise. How many times have we changed from our normal routine only to discover something, or someone wonderful, in a chance meeting or serendipitous connection? These moments can be designed into our lives. For more on this topic, go here to learn more.

Edge Compute will be more prevalent in lower cost devices: Not all products are connected to the internet at all times, and this can make AI harder to deploy on the product itself. AI on the edge means that much of the computer processing happens locally and not on a central server. This requires local processing power, more compact software, and requires more battery power (ie bigger battery, or shorter life). Advances in programming, chipsets, and battery technology will eventually overcome this, but for now, smart design can help limit many of these obstacles.

Edge compute has the benefits of faster processing time (lower latency) and the advantage of operation without internet connectivity. Imagine a blind person who relies on technology and mapping to navigate a busy New York street and imagine the challenges if the time delay in processing position, developing contextual awareness, and then providing input to the user has to travel from their device, up to the cloud, be computed, and then sent back to the user! While cloud systems are fast, edge commuting is faster giving this blind user much more immediate navigational feedback. When The Walt Disney Company created the Magic band, they required the user experience to be immediate. Project engineers then developed a technology that would connect to local computers to help park goers enter rides, unlock hotel doors, and pay for meals instantly. There’s an independent case study written on the subject you can find here.

AI can make us expert without becoming an expert: Just like well-designed surgical tools help all surgeons deliver similar outcomes, there are many things in life we’d like to be good at- cooking, drawing, learning a new language, being more social, being more artistic, doing our own simple repairs, putting on make-up with a more creative flair, or simply driving in traffic with less stress- these are all possible and AI will augment our own abilities so we can perform more competently without the training of being one.

AI can guide us physically, emotionally, and intellectually, but it has to be done in such a way that allows us to be unincumbered with the support AI can give us. Imagine wanting to copy the make-up application of someone you see online or Instagram and not feeling confident you understand the colors, products, and how to best apply. Now imagine you have a simple smart make-up pencil that not only can change color based on the Instagram phone, but has sensors and an automatic tip that can accurately apply the make upon your face like a professional. This combination of intelligence and devices is the future of consumer products- products that can help us achieve outcomes we aren’t able to do on ourselves, or do faster and better with a little AI help.

AI and E-commerce | Match made in Heaven: It should go without saying that if brands who deeply know their customers will be able to sell them more, and more relevant products. There are many devices and apps on the market that initially had no ecommerce back-end. L'Oréal for example acquired Modi-face in 2018 based on technology created at the University of Toronto, the AI/AR smart phone app that allowed users to virtually try on various make-up product to see how they looked when applied. Users could try different lipsticks, eyeshadow and foundation and blush colors to find just the right look. It took years however for this amazingly engaging app to stitch together the e-commerce back-end that would let these highly engaged users actually buy something.

Look for the use of AI to help product suppliers learn individual habits, preferences, and real-time emotional states to infer new marketing opportunities. Interestingly, I created a video demonstration of how AI can be used for beauty companies to collect real-time user data. The video demo can be seen here. Further, Adrich , a Pittsburgh-based start-up recently launched a small electronic tag complete with sensors that will indicate when and how a product, like skin cream or deodorant is used generating useful consumer insight data. You can find them here.

Data, Data- who’s got the data? AI and data go hand-in-hand. AI is built from data, and data feeds AI systems to be smarter (machine learning systems). Not only does data help AI to be smarter, it can help us marketers make better decisions, too. Let’s look at some examples of how AI enabled systems can be used to provide key insights to both users and product providers. Referencing the Magic band from Disney, not only did the band provide a friction-free park experience, it delivered massive amounts of data about where people went, what they ate, what rides they enjoyed, and when they used their hotel. Never before could such data be captured and harnessed to help designers improved rides, operations personnel better product ride usage, restaurants predict consumption, and see behaviors based on weather patterns for example.

Collecting data is one thing, turning all this data into actionable insights is yet another. Unilever, the company that owns such well-known brands as Dove, Knorr, Lipton, and Hellman’s, for example, built the “People Data Center” to collect data from around the world at various points in the user journey across the globe and 27 different categories. Data was drawn for sales, customer service inquiries, web searches, and cookie data (which is now not possible in all parts of the world) to help the company better understand consumer trends, competitors and how customers want to interact. You can learn more about Unilever ’s PDC in the Harvard Business review article here.

“The most successful companies don’t just have good products and strong distribution systems — they have a deep understanding of customers.” — Harvard Business Review

Without a digital presence, understanding consumers at scale is a challenge at best. Optimizing new product innovation, strategy, operations, marketing, service, and supply chain will rely heavily on data collection and those companies without a digital strategy will be left behind.

So, how do Consumer Products companies embrace AI? I’ve designed a simple rubric to help developers create better consumer product systems. But don’t be fooled by the simplicity of the approach- this is hard stuff to get right and frankly, many consumer product companies have difficulty executing on this simple objective. While critically important, this goes well beyond creating a good user experience and involves disciplines in AI, contextual awareness and behavioral neuroscience. In 2023, these types of skills should be table steaks for CP companies, but they’re not.

Here is a simple rubric (ICUS) for companies to follow to develop truly breakthrough products and services.

“I.C.U.S. is the acronym that will drive the development of useful AI enabled products in the next decade”

"Intent"- Develop the system that can understand the user intent and adjust accordingly to that intent. Think predictive modeling and temporal analytics (intent and actions over time). Services that are only as smart as the current transaction will fail.

"Context"- Understand the context in which the product is being use and who is using it. Sensors, situational context, and user awareness will be key. There's not much excuse now not to include multiple channels of data to paint a more holistic picture of the users needs in context.

"Usability"- Make the product seamless to use and anticipatory of the users needs. How many times have you felt stupid using an app and thought to yourself, “I’m a smart person, why can’t I figure out how to use this app?” 

"System"- Having had experience developing intelligent products and services leveraging sensors, AI, and a comprehensive digital ecosystem that open new revenue pathways, developing the skills organization, and pathways to create a truly friction-free experience requires a broad system view. Product, supply chain, customer service, data, data security, and data analytics must be designed together as a system from the start.

Where are we headed? We’re currently in a point of inflection where the technology isn’t yet ready to fully deliver on the promise or the expectation. AI systems will get smarter and more reliable than their predecessors of today and a whole new form of product and product experience will emerge beyond what we can even fathom today. Technology can change consumer behavior. A few years back however, did you ever think you’d be going to Walmart for healthcare? Or getting into a perfect stranger’s car for a ride? In the end, companies who can deliver a clear and unique proposition, with a consumer experience that is friction-free, solving a consumer need with a digital connection to its customer will win. 

About the Author: Eric is a recognized leader in consumer and health technology with more than 30 years of experience developing products that create new consumer and healthcare experiences using new technologies and human-centric approaches. Currently at Capgemini, he’s enjoyed a successful career in new product innovation, as a healthcare company founder, publishes a monthly podcast Interesting People for 20, serves as a mentor at Harvard Business School, and runs Otis Venture Partners where he advises and supports early-stage founders. Eric invented the Reebok Pump, and was instrumental in creating many consumer and medical products on the market today. Eric is a competitive cyclist and holds a degree in mechanical engineering.

That’s an awesome rubric and still very applicable

Like
Reply

To view or add a comment, sign in

More articles by Eric Cohen

Insights from the community

Others also viewed

Explore topics