Any sufficiently advanced technology is indistinguishable from magic.
Arthur C. Clarke
AI has captured our imagination and shattered all technology adoption records.
Each month, it raises the bar on what it can accomplish. We have moved from being bewildered to holding our collective breaths and bracing for unknown unknowns.
The question isn’t if AI will integrate into how we work, live and play – it is a foregone conclusion that AI will become a public good, like electricity.
A public good like electricity is affordable, present in every household and so deeply embedded into our lives that we take it for granted.
This journey to today has not been easy. Electricity and its twin, the Industrial Revolution, might have felt inconvenient at first. Even dangerous.
People got laid off or were forced to upskill (or else), entire factories had to be reassembled or built from scratch. Sadly, because we did not know how to handle its power, many people even got electrocuted and died.
This did not stop the steady march of technology and before we knew it, how we worked and lived, what we consumed and spent time on had changed forever.
Over time, electricity set many flywheels in motion.
Here are three obvious ones.
electricity → Industrial Revolution → more efficient production → more and better products at lower prices → marketing → more wants → more production
electricity allowed us to study, write, and create well into the night → longer productivity window → more ideas → more inventions → more technologies → more leisure time.
electricity → more productivity → more leisure time → more creation of art → more appreciation of art → more culture → more ‘civilisation’.
Technology by itself is not bad. People use technology badly. Have people not used electricity for evil? 100% they have.
Yet, at the end of it all, when we stack up the first, second, third and hundredth order effects of electricity, of will say this with surety. Net-net, we are better off.
So I wanted to imagine a future with AI as a public good – AI will be as normal, as invisible, but much much more powerful than electricity.
We will end up with net positive outcomes only if we negotiate the existential dilemmas AI will throw at us.
AI will become a public good like electricity. But we will be net-net better off only if we negotiate the existential dilemmas AI-powered ecosystems will through at us.
AI will not just changes jobs, or make us more efficient. It will change the fundamental structure of value, differentiation and business models.
Today, I look at scenario #1.
- Business models, value, and differentiation will reside at the convergence of technologies, products, data and services.
- We will live in a two-sided AI agentic network economy, powered by data – like a buy-sell-stock market.
- The dilemmas will be around value capture, brand power and data hoarding.
Business models, value, and differentiation will reside at the convergence of technologies, products, data and services.
Electricity is a transversal technology – it penetrates other products and technologies to create value that is bigger than the sum of its parts.
For example, a table lamp is just a piece of metal without electricity coursing through it and electricity is wasted without a vehicle to channelize its energy.
Businesses and consumers are in the business of value exchange – products or services in exchange for money and goodwill. Yet, both are starved for meaningful differentiation.
In the age of AI, convergence of technologies, products, data and services will create new vectors of value and differentiation, which will forever change business models.
Here are two examples.
We will go from linear one-at-a-time transactions, e.g. buy hair color in exchange for money, to messy, many-at-a-time bundled transactions
If you are finding this interesting, chances are your friend will too.
Hair color as a service
Tomorrow, a hair color product alone will offer lesser differentiated value than a product-service-data bundle of L’Oréal, Urban Company and Lal Path Labs.
Through a Lal Path lab test I will know my hair texture, hair health and scalp chemistry → L’Oréal will use this data to personalise a hair color formula for my scalp → This will be delivered and administered by Urban Company in the comfort of my home.
Travel as a product
Any trip involves hours of coordination and administrative work.
Tomorrow, a trip might involve a bundled offering of Uber, Trip Advisor, iPhone, Make My Trip and a Taj Hotel Chef.
Trip Advisor will curate an ideal itinerary (cafes, hotel, co-working spaces) after matching reviews from lookalike audiences based on my preferences, interests, location, types of meetings, trip goals, and health goals → Once I approve it, automated execution will kick in → Make My Trip ticketing, check-in right upto Digiyatra → Hotel and cafe reservations will be made → A self-driving Uber will adjust for real time traffic and will show up at my doorstep just in time → my iPhone alarm clock will be adjusted accordingly → And the delight to top it off? Taj chef ready with a menu of local cuisine adapted to my health goals, at my destination.
We will live in a two-sided AI agentic network economy, powered by data- like a buy-sell stock market.
Each of us will have a personal AI agent
My data will reside with my personal AI agent.
It will be in my interest to train my AI agent to deeply understand my likes, dislikes, goals, and interests. Over time, it will not just anticipate my choices, but also optimise them for the best outcomes for my goals.
My AI agent becomes a consumer market of one.
Different consumer markets of one will have different goals. One might want to maximise health, another to save time yet a third might be a discount seeker.
On the other side, companies will have their own AI agents.
These agents will be trained on each company’s business model, product benefits, production constraints, brand positioning, and goals.
Each of these is a company market of one.
Each company AI agent will interact with millions (maybe billions) of consumer markets of one and personalise service-product bundles for them1.
My AI agent could engage with each company’s AI individually and stitch together a product-service bundle for my goals.
Alternatively, my agent could to work with one lead agent, e.g. trip advisor, and the trip advisor agent can engage with all the others and offer one bundled offer to my agent.
Maybe whichever product or service is first in my consumer journey, will become the lead agent from the company side.
In both scenarios, think of company AI agents as sales persons cum brand custodians supported by data trends, empowered to collab with other company AI agents and to optimise the best outcomes for the two-sided network system.
Imagine an underground stock market of AI agents negotiating, trading, bartering, bullying each other to get the best deal for their consumer market of one and company market of one respectively.
Seems like magic doesn’t it?
Arthur C. Clarke would agree. This is magical for consumers, but would require contortions of a world class gymnast from businesses.
If AI agents play the long game, they will optimise outcomes for all parties. On the other hand, short-term mindsets will lead to an air of mistrust and win-lose commercial wars.
AI agentic algorithms will navigate three critical dilemmas.
The dilemmas will be around value capture, brand power and data hoarding.
VALUE CAPTURE
Given that all companies involved will create value for the consumer market of one, who will capture the lion’s share of value?
Will it be the one with the strongest brand or the one my AI agent approached first?
Will it be the product with best NPS scores from me? Or the one I buy most frequently?
Or will value capture be a cold scientific equation of cost of goods sold + profit margin?
DATA SHARING
Since data algorithms will fuel all transactions, it becomes precious currency. Who will share data, how much and how?
If each consumer is a market of one, the more data each company knows about me, more they can enrich my experience. Will the company with the best data on me capture the highest value?
Whose data is it anyway?
BRAND
Will the stronger brand be the face of the entire product-service bundle for the consumer market of one?
Will the rest of the brands be white listed?
Will brand even matter?
In a world where cold hard data makes buy-sell decisions in microseconds, maybe brand trust will be built purely on performance? And emotional pay offs will have to be given numbers instead of words?
Or will brand become even more important because I will feed in my brand preferences while I train my AI agent?
There are more questions than answers right now.
Thanks for reading. And I will share another AI-powered future scenario next week.
More ways to connect
- Connect on: Linkedin
- Visit my: Website
- Subscribe to this Newsletter
Thanks for reading Performonks! Subscribe for free to receive new posts and support my work.
1 This will mean that performance marketing will target AI agents, not humans.