Generative AI Productivity is an Issue, But So is Most IT

Quantifying or documenting generative artificial intelligence value is a top issue, respondents said in a Gartner survey of information technology executives. That really should not come as a surprise, as documenting the value of most technologies in knowledge or office work is challenging. 

And since generative AI is used for customer service interactions, producing summaries, developing code, drafting documents or messages, the issue is how well we can document the productivity lift from virtually any IT tool, in those instances. 

source: Gartner 

Quantifying the productivity gains from new IT solutions in customer service can be surprisingly challenging, experts often say. As applied to customer service agent operations, IT tools are said to improve customer satisfaction, handle volume fluctuations, and reduce training times. But isolating the impact on individual agent output can be difficult.

Generative AI and other IT might increase the number of customer contacts per hour, for example. Chatbots are a substitute for human agents as well, so might contain customer service costs. But that all hinges on the quality of the chatbot to answer the questions customers actually have.  

In addition, customer service involves interactions with various channels (phone, email, chat), making it hard to isolate the impact of IT on a single metric. Improved customer satisfaction might not directly translate to a quantifiable reduction in call times.

But that might not always correlate with improved ability to actually solve a customer problem. In other words, quantity is not the same as quality. 

Also, changes in productivity may not be immediate.  Learning curves, process adjustments, and cultural shifts within the team can take time to settle before the full impact is realized.

Accurately measuring before-and-after states requires clean data and proper attribution. Factors like seasonal variations, changes in customer behavior, or external promotions can skew the results. 

Demonstrating a clear return on investment (ROI) for new IT implemented in customer service can be challenging. Here's a breakdown of the difficulties:

Generative AI also might not eliminate tasks, but rather shift them. Increased efficiency in handling routine inquiries might free up agents for more complex issues, making it difficult to show a direct reduction in overall workload or quality of outcomes. 

Improved agent morale, reduced stress, and better customer experiences are all positive outcomes, but they're not easily captured in traditional productivity metrics like call resolution times.

Source	Author(s)	Publication	Key Findings
The Impact of Information Technology on Customer Service Productivity	Brynjolfsson, Erik, et al.	Management Science (1993)	Found that the impact of IT on productivity depends on the specific technology and how it's implemented. Identifying productivity gains requires careful analysis.
Does IT Really Pay Off? Measuring the Effects of Information Technology Investment on Customer Service	Lee, Sang-Pil, and Byung-Il Park	Journal of Service Research (2001)	Highlights the difficulty of isolating the impact of IT on productivity due to the presence of confounding variables. Emphasizes the need for a multi-faceted approach to evaluation.
The Challenges of Measuring the Business Value of Customer Relationship Management (CRM) Initiatives	Rust, Roland T., et al.	Journal of Marketing (2004)	Argues that traditional ROI metrics might not capture the full value of CRM systems, which often include customer satisfaction and loyalty benefits alongside productivity improvements.

We might observe similar issues with other tasks GenAI might help with, such as creating documents and text. Past applications of word processing arguably provide speed and quality advantages that are hard to quantify. 

Study Name	Venue	Date	Key Conclusions
"The Effects of Word Processing Software on Writing Performance" by Keith S. McNeil	Educational Technology Research and Development	1988	Found that word processing had a minimal impact on writing speed but improved editing efficiency.
"The Impact of Technology on Writing: A Review of the Literature" by Charles A. MacArthur, et al.	Review of Educational Research	2001	Concluded that the impact of technology on writing quality is mixed and depends on factors like task and user skill level.
"The Myth of Increased Productivity: How New Technologies Slow Us Down" by Sherry Turkle	Basic Books	2015	Argues that constant connectivity and information overload can hinder focused work and deep thinking, potentially impacting writing productivity.
"Beyond the Efficiency Paradigm: Rethinking the Role of IT in Knowledge Work" by Wanda J. Orlikowski	Organization Science	2007	Shifts the focus from measuring just efficiency gains to considering how IT can enhance creativity, innovation, and collaboration in knowledge work like writing.
"The Myth of Increased Productivity: How New Technologies Can Slow Us Down" by Daniel H. Pink	Harvard Business Review	2016	Argues that constant connectivity and information overload can actually decrease focus and productivity, even with advanced tools.
"The Paradox of E-Mail: A Sociotechnical Perspective on Communication Overload" by Stefan Klein, Christian Bartsch, and Jan Marco Leimeister	Journal of Communication	2004	Highlights how email, a seemingly efficient communication tool, can lead to information overload and ultimately hinder productivity.
"The Impact of New Information Technologies on Task Performance: A Meta-Analysis" by Steven E. Fiore, James R. Salas, Michelle H. Cuevas, and Cheryl A. Bowers	Human Factors	2003	Reviews multiple studies on the impact of technology on task performance and concludes that the effect can be positive, negative, or neutral depending on the specific task and user characteristics.

Cable TV Was About Linear Programming Choice; Streaming Remains About On-Demand, "Live" Features Notwithstanding

One hears a lot of chatter these days that video streaming bundles are the new cable TV. No doubt, it is a substitute or replacement product for subscription TV, to a growing extent. And yet the allegation that streaming is “becoming cable TV,” often uttered as a negative, might miss important elements of the shift. 

Cable TV was initially a subscription-based way to watch linear broadcast TV for rural residents. Later it had appeal for metro-area residents as a “more choice” vehicle. But all those new and specialized content formats remained linear. 

On-demand consumption of movie and other pre-recorded content was not possible until the advent of the videocassette recorder and the business of video rentals. In one sense this also was an extension of the “more choice” shift of consumption patterns. 

But that arguably was secondary. The primary shift was from linear to on-demand consumption, a trend that continued with the availability of DVDs and then video streaming services. It is not insignificant that Netflix began life as a video rental retailer before making the shift to streaming delivery. 

On the other hand, streaming services are adding more exclusive “live” content as well, largely because of the customer interest in sports programming, for example. Such content aids customer acquisition and retention in the same way that other exclusive content creates differentiation and value. 

All that has financial implications for broadcast, cable TV and video streaming industries, as broadcast TV and cable TV advertising revenue shifts to other venues; cable TV subscriptions decrease and streaming subscriptions increase. 

 

Year	Broadcast TV Advertising Revenue ($B)	Cable TV Advertising Revenue	Cable TV Subscription Revenue	Video Streaming Subscription Revenue	Video Streaming Ad Revenue
2023 (Estimated)	40	60	120	18.8	4
2024	38.5	58	115	22.5	4.5
2025	37	55	110	26.5	5
2026	35.5	52	105	31	5.5
2027	34	49	100	36	6
2028	32.5	46	95	41.5	6.5
2029	31	43	90	47.5	7
2030	29.5	40	85	54	7.5

On the other hand, customers can use DVR capabilities to create a sort of on-demand capability for linear broadcast and cable TV programming as well. 

The point is that streaming video bundles are not the "new cable TV." Both cable and streaming are about content choice. But all forms of broadcast TV are essentially linear, with a DVR overlay. Streaming bundles remain an on-demand value proposition; the latest evolution of on-demand video. That remains true even as streaming services add some "live" programming as well.

Will All PCs Eventually be AI PCs?

It’s arguably too early to be confident about market share forecasts for AI PCs. Will they eventually become the standard PC, as smartphones now are the standard phones purchased by consumers? Or will AI PCs remain specialist tools, to a greater or lesser degree?

The issue is less “ability to use AI” and more the issue of the value of local processing. 

One might liken the market prospects to that of consumer-grade or work-grade general-purpose PCs and “workstations” used by some, but not most. 

High-end workstations are used for 3D rendering, video editing, and complex simulations. They prioritize raw local processing power, high-performance graphics cards and large memory capacities. 

AI PCs will feature specialized hardware components like Tensor Processing Units or Neural Processing Units alongside traditional CPUs and GPUs to accelerate AI computations. But that does not necessarily speak to “why” such machines would add value over PCs not including those elements. 

At least for the moment, forecasters see a gradual shift of buying patterns. 

Year	AI PC Market Share (%)	General-Purpose PC Market Share (%)	Sources
2024 (estimated)	2-5	95-98	Canalys, Grand View Research
2025	5-10	90-95	Canalys
2026	8-15	85-92	Gartner
2027	12-20	80-88	Gartner
2030	20-30	70-80	IDC

AI PCs might be useful for developers or users working on AI training, inference and other AI-dependent workloads. For most consumers, that might include image processing, speech-to-text,  language translation or gaming. 

Some professionals who are AI developers, researchers or data scientists might have work-related reasons that make AI PCs a good choice, if local processing adds value, compared to remote processing. 

It is not clear how much of the video editing or 3D rendering market might be affected. “Professional” use cases might not be supported, but casual and user-generated content might be. 

There arguably is more debate about the PC market than the smartphone market, though. AI already makes general sense for image processing on smartphones as well as speech-to-text. But additional use cases requiring on-board processing will have to be developed. 

The argument for local AI processing on PCs is more complex. AI could personalize software functions, optimize battery usage, or enhance security measures. But it is not certain those tasks must be handled by on-board processors.

AI-specific hardware could significantly improve device performance for tasks such as photo and video editing, gaming, or augmented reality applications, to the extent those features are deemed useful on PCs. 

Battery life might be a constraint for smartphones and laptops, though. And, for most users, additional AI device cost will have to be balanced against “new” and valued use cases. It will take some time for those use cases to develop. 

At least in principle, one might envision a new category of AI PCs half way between workstations and general-purpose PCs. One might also envision an eventual migration of local AI processing to most PCs as a regular feature, at some point. 

Are Large Language Models Investor "PIcks and Shovels" or Not?

Aside from all else, artificial reality Iis an investment theme. A study by Morgan Stanley, for example, argues that AI's materiality to investment theses has increased significantly, affecting at least 446 stocks, worth $15 trillion, in 2024. 

That is logical enough, given the importance of AI enabling technology including semiconductors, servers, cloud computing, data centers, energy sources for data centers; as well as AI implications for the functionality of enterprise and consumer software. 

In 2023 and 2024, much of that financial impact has centered on generative AI, eclipsing machine learning and natural language processing, for example, which already are used by many business and consumer applications. 

Facial recognition, for example, uses ML algorithms to unlock user smartphones, while digital voice assistants such as Siri and Alexa use AI, NLP and ML to understand commands and carry out a range of tasks. 

AI algorithms are used in e-commerce to make personalized shopping recommendations; in clinical trials to improve drug discovery and efficiency and elsewhere across an array of industries to automate a host of back-office tasks.

As often is the case, suppliers of “picks and shovels” were among early winners. Though we might quibble about what firms are in that category, many would say suppliers of graphics processor units, acceleration chips, memory, cloud computing suppliers and even electrical power companies are among firms and industries supplying “picks and shovels.”

There arguably is greater disagreement about whether large language models are enablers--and therefore in the “picks and shovels” category--or in the many other categories of beneficiaries of AI. 

Category	Industries/Firms	Description
AI Enablers (Picks & Shovels)	Chipmakers (Nvidia, AMD, Intel) Cloud Computing Providers such as Amazon Web Services, Microsoft Azure, Google Cloud Platform)   Data Labeling Companies , Labelbox, Scale AI)  OpenAI (research & development) Electrical Utilities	Foundational infrastructure and tools needed to train and develop AI models in general. They don't necessarily focus on specific applications of AI.
Generative AI Beneficiaries	Content Creation (Adobe, Unity, Unreal Engine)   Drug Discovery (Insilico Medicine, Generative Bio)   Materials Science  Marketing and Advertising (Anyword, Copy.ai) Social Media Search Business Services Law Transportation Retailing Information Technology	Industries and firms leverage generative AI for specific applications. For example, generative AI can create new marketing copy, design elements, or discover new materials.

Large language models are where large amounts of disagreement could occur. Large language models can be viewed as both beneficiaries of AI and, to a lesser extent, enablers (picks and shovels). GPUs, memory, power, data centers and cloud computing enable AI to run. 

In that sense, AI platforms and apps are beneficiaries, not picks and shovels. LLMs are a product of advanced AI techniques such as deep learning and natural language processing, but are applications in their own right. And AI applications are not generally considered to be picks and shovels. 

Still, many could argue that LLMs are enablers to the extent they support creation of software features and applications. LLMs will power search, social media, many forms of app personalization, smartphone image processing, speech to text functions, text summarization, image processing, notetaking and so forth. 

So LLMs are where the distinction between AI “picks and shovels” enablers and and AI beneficiaries is mixed. 

Further muddling exists because some beneficiaries of AI also are developers of LLMs, and are revenue generators. Think of Microsoft’s Copilot, Google’s Gemini or other GenAI apps sold as subscriptions. GenAI is available both as a feature of Microsoft and Alphabet (Google) products as well as a subscription-based application. 

In some instances the LLM is an enabler or feature; in other cases an app. For some firms, GenAI and LLMs are both enablers (picks and shovels) and beneficiaries of capabilities and features offered by most products and processes. 

That matters for equity investors as well as all sorts of firms, industries and products.