AI Assistant Revenue Upside Mostly Will be Measured Indirectly

Amazon expects Rufus, its AI shopping assistant, to indirectly contribute over $700 million in operating profits this year, Business Intelligence says. 

The expected upside would come in the form of "downstream impact," a metric Amazon uses to measure a product or service's potential to generate additional consumer spending across Amazon's vast offerings. Rufus, as such, generates no direct revenue, of course. 

Rufus product recommendations might lead to more purchases on Amazon's marketplace, for example. The value of advertising embedded in Rufus content are another way indirect revenue upside is measured. 

By 2027, however, it is expected to reach $1.2 billion in DSI profit contributions, according to Amazon. 

“From broad research at the start of a shopping journey such as ‘what to consider when buying running shoes?’ to comparisons such as ‘what are the differences between trail and road running shoes?’ to more specific questions such as ‘are these durable?’, Rufus meaningfully improves how easy it is for customers to find and discover the best products to meet their needs, Amazon says. 

That is likely a way most firms are going to have to rely upon to quantify their LLM assistant revenue gains. 

Use Case	Description	Revenue Impact
Customer Support Automation	AI chatbots handle FAQs and troubleshooting, reducing customer service costs.	Lowers operational costs and improves customer retention.
Lead Generation,  Qualification	AI assistants engage website visitors, collect data, and qualify leads.	Increases conversion rates and enhances sales pipeline efficiency.
E-commerce Upselling,  Cross-Selling	AI recommends relevant products based on user behavior and preferences.	Boosts average order value and sales.
Content & SEO Optimization	AI generates blog posts, product descriptions, and metadata for SEO.	Increases organic traffic, improving brand visibility and sales.
Personalized Marketing, Retargeting	AI-driven chatbots deliver personalized offers and recommendations.	Enhances engagement, conversion rates, and repeat purchases.
Employee Productivity Enhancement	AI automates repetitive tasks (e.g., email drafting, summarization, scheduling).	Saves time, allowing employees to focus on high-value tasks.
Market Research,  Insights	AI collects and analyzes customer feedback for business insights.	Improves decision-making and product-market fit.
Training, Onboarding	AI-based interactive training modules for new employees.	Reduces onboarding time and training costs.
Subscription, Membership Services	AI chatbots engage users to promote premium subscriptions.	Increases subscription revenue and customer lifetime value.
Reducing Churn,  Customer Retention	AI proactively engages users before they disengage or cancel services.	Lowers customer acquisition costs by improving retention rates.

If Generative AI is "Winner Take All," That Dictates Investment Bets

Contestants in the generative artificial intelligence model business are following a “winner takes all” approach to the market, investing at a pace that gets criticized by financial analysts (and rightly so, in some respects) for exceeding what seems immediate and tangible financial payback.

But the “winner takes all” strategy has worked often in the internet era, across many different segments of the market. So we already can predict some outcomes. 

“Wild” levels of spending are going to pay off for at least one and perhaps two providers of GenAI models, in terms of leadership of the market (ecosystems built on them). Some of the contestants will eventually “break even” for their investors, neither gaining or losing equity value. 

But most will fail, losing their investors most to all of the invested capital. That is simply what has happened in winner-take-all markets. 

As we saw early on, even pre-internet, for operating systems, and later saws for search engines, mobile operating systems, e-commerce, social media, ridesharing, peer-to-peer lodging and other app segments, market leadership is highly concentrated with one or sometimes two providers dominating. 

So the companies developing GenAI frontier models, such as OpenAI, Google DeepMind, and Anthropic, are assuming the market  will eventually shake out with a “winner takes all” structure, with market leadership highly concentrated and creating a new ecosystem of value around the leading models, with the hundreds of other would-be contestants relegated to history.

Market/Product Category	Dominant Players
Search Engines	Google
Social Media	Facebook, Instagram, TikTok
E-commerce	Amazon
Mobile Operating Systems	Apple iOS, Google Android
Desktop Operating Systems	Microsoft Windows, macOS
Web Browsers	Google Chrome, Safari
Streaming Video	Netflix, YouTube
Cloud Computing	Amazon AWS, Microsoft Azure, Google Cloud
Online Advertising	Google, Meta (Facebook)
Smartphones	Apple (iPhone)
Productivity Software	Microsoft Office, Google Workspace
Mobile Payment Systems	Apple Pay, Google Pay, PayPal
Digital Maps	Google Maps
Food Delivery Apps	DoorDash, Uber Eats
Ride-Hailing Apps	Uber, Lyft (US), Didi (China)
Video Conferencing	Zoom, Microsoft Teams

By way of contrast, end users of GenAI--especially enterprises--will have to answer for their choices, as they would for any other capital investments made to support their businesses. And that goes for suppliers fo hardware and software incorporating AI functionality. 

GenAI model leadership will undoubtedly shake out into a “winner takes all” pattern. But that also means hardware or software suppliers might “bet on the wrong horse,” as might their customers. That generally means high danger for model suppliers and those firms that are direct parts of those ecosystems. 

Hardware and software participants might face moderate to high impact if they are in the “wrong” ecosystem or platform. End users--whether consumers or businesses--might face low repercussions if their chosen GenAI supplier does not emerge as the eventual market leader (either provider one or two). 

Basic or firm-specific functionality might not be the big problem, as niche solutions might be perfectly viable, especially if a model emerges as providing superior industry-specific functionality. 

Value Chain Segment	Example Participants	Degree of Danger if Supplier/Partner is Not a Winner	Reason for Risk
Foundation Model Providers	OpenAI, Google DeepMind, Anthropic, Meta	High	If the chosen provider loses, end-users face high costs in switching models due to integration dependencies.
Model Fine-Tuners & Adaptation	Hugging Face, Scale AI, Stability AI	High	Fine-tuning investments and adaptations may not be transferrable to new, winning models.
Hardware Providers	Nvidia, AMD, Intel, custom chip startups	Moderate to High	Dependence on specialized chips may limit flexibility; non-dominant players may face R&D cutbacks.
AI Infrastructure Platforms	AWS, Microsoft Azure, Google Cloud	High	Non-winner platforms may lack interoperability, leading to high switching costs and stranded data assets.
Middleware Providers	Databricks, Snowflake, API integration services	Moderate to High	Middleware built for a specific provider may need reconfiguration, reducing compatibility with other models.
Application & Service Developers	Jasper, ChatGPT plugins, custom AI tools	Moderate to High	Apps tightly integrated with a non-winning model may struggle with compatibility and require costly rewrites.
Consulting & Implementation	Deloitte, Accenture, AI-specific consultancies	Moderate	Models selected may lose support, limiting longevity of solutions delivered to clients.
AI Tooling & DevOps	MLOps tools, AI pipeline solutions (e.g., Weights & Biases)	Moderate	DevOps tools tied to a specific model may lose relevance if clients pivot to dominant providers.
Enterprise & Custom Solutions	SAP AI, Oracle AI, Salesforce Einstein	Moderate	Custom solutions based on non-winning models may have compatibility and feature limitations.
Data Providers	Web-scraping firms, domain-specific datasets	Low to Moderate	Dependence on any one model is less critical, though dataset compatibilities may shift.
Consumer-facing Applications	Grammarly, Canva, AI-powered image and video tools	Low to Moderate	User-facing tools may need updates for compatibility but can adapt faster than core infrastructure layers.
End Users & Businesses	Large enterprises, SMEs, individual users	Low to Moderate	Users may face inconvenience, but model-agnostic interfaces could minimize direct switching costs.

Still, longer term, the specific GenAI ecosystem should matter, as network effects should eventually emerge.

Do Home Broadband Speed Rankings Really Matter Much?

Ookla’s May 2024 report on mobile and home broadband shows Singapore and Hong Kong leading the list of countries with the fastest speeds, which is not surprising at all. 

You might not have expected Chile to rank third, the UAE and Iceland in spots four and five. The United States ranks sixth, which is sort of an anomaly. Over the past half century or so, it would not have been uncommon for U.S. metrics to rank anywhere from 12th to 20th on measures of tele-density or internet access bandwidth. 

We might reasonably ask how much importance such speed rankings actually mean. One might argue the rankings generally suggest that small city-states and small countries can produce good broadband infrastructure faster and better than any large country, simply because the physical facilities are smaller in coverage area, with higher density. And network size and population density directly affect the cost of such facilities. 

Hong Kong, Singapore and other such areas will always be able to create high-performance access infrastructure faster than any continent-sized country with low population density. 

Nor, looking only at city-states and small countries, might we see clear correlations between growth and home broadband speeds. Singapore and UAE might be strong performers in that regard. But other small countries might not show the same strong correlations. It might be the case that only rarely, if ever, are home broadband and economic growth rates uncorrelated. 

But the correlations are not consistent. So it is worth speculating about how important such rankings actually are, when it comes to applying the tools and wringing business or economic value out of them. 

To be sure, lots of studies suggest there is a correlation between economic growth (gross domestic product) and home broadband availability and speed, with perhaps greater correlations related to availability than speed. 

Study	Year	Key Findings
Ericsson, Arthur D. Little, and Chalmers University of Technology	2011	Doubling broadband speeds can add 0.3% to GDP growth
World Bank	2009	10% increase in broadband penetration associated with 1.38% increase in GDP growth for developing countries
OECD	2011	Positive but diminishing returns from increased broadband speeds on economic growth
ITU (International Telecommunication Union)	2012	Broadband has a statistically significant impact on GDP growth, but effect varies by region and level of development
Rohman and Bohlin	2012	Doubling broadband speed contributes 0.3% GDP growth in OECD countries

But it might also be worth noting that there are similar correlations between gross domestic product gains and educational attainment; rule of law; capital investment; income and wealth; or infrastructure density and availability. 

And correlation is not causation. 

In fact, “causality” might even be the reverse of what we might think. 

Keep in mind that economists generally economists might generally agree there is a  “causal” relationship between growth and:

• Capital accumulation (both physical and human)  

• Innovation and technological progress (research and development; creation of new ideas)

• Macroeconomic stability helps (Low and stable inflation; sound fiscal policies)

• Openness to trade

• Quality Institutions (rule of law and low levels of corruption)

• Financial markets well developed

So we might consider education an input to future capital; innovation or technology development. We might consider home broadband another form of capital. 

But it's often unclear whether some factors said to cause growth are themselves caused by growth. Does financial development, trade openness and political stability cause growth, or does growth cause financial development, trade and political stability? We cannot really say. 

Consider “good schools,” quality home broadband, medical care or other supposed platforms aiding growth. 

It might plausibly be the case that demand for good schools and fast internet access, for example. Are the product of demand from citizens who already have the resources to pay for such quality broadband, as well as the use cases. 

Likewise, if local schools are funded by property taxes, then “good schools” might be “caused” by affluent citizens who can afford expensive housing, which comes with high property values, leading to high tax revenues to fund schools. 

In fact, one might well argue that often, the prevalence of quality home broadband, transportation infrastructure or any number of other supposed producers of economic growth might instead be a result of pre-existing strong economic growth. 

Rather than robust economic growth being “created” by quality broadband; educational attainment and other drivers, it is equally plausible that pre-existing high growth creates wealth and resources that in turn lead to the other outcomes. 

You might suspect educational attainment, for example, is correlated with stronger economic growth, and studies support that notion. But a flywheel might be at work, where pre-existing high attainment leads to more attainment; high growth reinforcing more high growth. 

Study/Source	Correlation/Finding
Georgia Tech study	0.75 correlation between years of education and GDP per capita. 1 year increase in education associated with 34.4% increase in GDP per capita.
Hanushek & Peterson analysis	Raising US student test scores to Canadian levels estimated to add $77 trillion to US economy over 80 years.
International comparison	Countries with top test scores (e.g. Singapore, Hong Kong) had ~2% higher annual GDP growth compared to average.
OECD countries analysis	Positive correlation between education expenditure at all levels and GDP, stronger over 5-10 year periods.
Developing countries analysis	Positive correlation between primary education spending and GDP growth. Negative correlation for secondary/higher education.
General finding	Education is "intrinsically linked to economic growth", influencing both personal salaries and national GDP.

Likewise, studies of transportation infrastructure also tend to be correlated with gross domestic product, but sometimes only moderately. 

Transportation Mode/Metric	Correlation with GDP	Time Period	Source/Study
Civil aviation (freight)	0.907 (high)	1990-2007	IOP Science study
Civil aviation (freight)	0.711 (strong)	2008-2017	IOP Science study
Inland waterway (freight)	0.816 (strong)	1990-2007	IOP Science study
Inland waterway (freight)	0.789 (strong)	2008-2017	IOP Science study
Road transport (freight)	0.715 (strong)	1990-2007	IOP Science study
Road transport (freight)	0.741 (strong)	2008-2017	IOP Science study
Railway (freight)	0.668 (strong)	1990-2007	IOP Science study
Railway (freight)	0.558 (moderate)	2008-2017	IOP Science study
Water transportation (freight)	0.750 (strongest)	1989-2018	E3S Conferences study
Highway (freight)	0.709 (strong)	1989-2018	E3S Conferences study
Pipeline (freight)	0.700 (strong)	1989-2018	E3S Conferences study
Railway (freight)	0.678 (strong)	1989-2018	E3S Conferences study
Civil aviation (freight)	0.593 (moderate)	1989-2018	E3S Conferences study

The point is that we cannot be very sure that faster home broadband is the result of growth or the cause of growth. Nor can we know very much about how the “quality” of broadband (speed and latency performance, for example) produces growth or is a reflection of growth.