Generative AI Already Shows Value for Coders

Perhaps quietly, generative artificial intelligence is going to show return on investment in software development. Some of that gain occurs when coders can use GenAI to check their code. 

Perhaps the greatest upside comes when code development can be automated. Most estimates suggest time savings between 20 percent to 50 percent in code creation, with related cost savings between 10 percent to 25 percent in labor costs. 

Impact Area	Estimated Time Savings	Estimated Cost Savings
Code Generation	20-50% reduced coding time	10-25% reduced labor costs
Code Review	30-50% reduced review time	15-30% reduced quality assurance costs
Bug Detection	20-30% reduced debugging time	10-20% reduced maintenance costs
Documentation	40-60% reduced documentation time	20-30% reduced knowledge transfer costs
Learning Curve	20-30% reduced onboarding time	10-20% reduced training costs

Separately, GenAI is a potential disruptor of existing software firms, for the same reasons GenAI saves time and money for code development.

Also, as GenAI is a potential disruptor of existing business models in a range of content-based industries including search, question-and-answer sites, news and information, video and audio entertainment, it might also be the foundation for disruptive attacks on legacy software markets. 

Some believe Adobe Acrobat could be disrupted by GenAI content creation as an alternative, for example. GenAI also is seen as a threat to search, question-and-answer sites and parts of other content-creating businesses ranging from video and audio entertainment to news and information sources. 

Study	Date	Publisher	Key Conclusions
The Economic Impacts of Generative AI on Software Development	2023	McKinsey & Company	Potential 2540% reduction in software development costs  2030% increase in developer productivity
Generative AI in Software Engineering: A Productivity Analysis	2022	IEEE	Up to 30% reduction in time spent on coding tasks  1525% overall cost savings for software projects
The ROI of AIAssisted Coding	2023	Forrester Research	2035% decrease in bug fixing and maintenance costs  $515 million annual savings for large enterprises
Measuring the Impact of GitHub Copilot on Developer Productivity	2022	GitHub	55% faster code completion  3040% reduction in time spent on repetitive coding tasks
AI and the Future of Software Development	2023	Gartner	40% of application development will use AIassisted coding by 2025  1530% potential reduction in development costs

Could Android or Chrome Survive or Thrive as Independent Companies?

“Never let accountants or lawyers run your business,” some of us tend to believe, as such professions are, by design, risk averse, and business growth often requires embracing some amount of risk. So perhaps only lawyers could come up with the idea that separating Android or Chrome from Google search is a workable idea.

Some of us would probably point out that neither Android nor Chrome actually produces much direct revenue--if any--for Alphabet or Google, and it is questionable whether either product, if owned by new and standalone companies, would be able to sustain a profitable business model very easily, if at all. 

How many firms--if any--would be able to invest in and sustain a global and “free” smartphone operating system if it is required to create a new revenue model because Google does not subsidize it? One might ask whether revenue earned by the Google Play Store would be sufficient to sustain Android. 

By some industry estimates the Google Play Store generated between $40 billion to $50 billion in 2023, from app sales, in-app purchases, subscriptions and ad revenue from apps distributed on the platform.

As with any retailer, Google earns a commission or percentage of such sales. Google typically takes a 30 percent share of app sales and in-app purchases (reduced to 15 percent for developers making under $1 million per year), amounts similar to Apple’s App Store model. 

Google also earns 15 percent on subscriptions after the first year, encouraging developers to build long-term subscription models. 

So if the Google Play Store generated $40-50 billion in 2023, Google’s share would likely be in the range of $12 billion to $15 billion. Assuming nothing else were to change, any independent Android company might hope to hang on to most of that revenue stream. 

But if independent Android and Chrome companies were created, there is no assurance all else would remain the same, as the Google app and device ecosystem could very well change if the operating systems and browser were no longer integrated within the broader Google application ecosystem. 

Consider costs for Apple to support iOS, the equivalent to Android. 

Apple employs thousands of engineers and developers to work on iOS. Salaries, benefits, and overhead for these teams could total between $750 million to $1.5 billion annually.

Apple’s total cloud spending is estimated to be around $4 billion to $5 billion per year, with a portion allocated specifically to iOS services.

Apple invests heavily in marketing to promote new iOS features, updates, and devices. Apple’s annual marketing budget is estimated to be around $2 billion to $3 billion, with a significant portion allocated to iOS.

Apple provides customer support for iOS devices, which includes call centers, online support, and in-store assistance. This could range from $500 million to $1 billion annually.

Apple’s overall research and development spending was around $27 billion in recent years, with a significant portion directed toward iOS development.

Based on those assumptions, Apple iOS costs might range from a low of $4 billion annually up to perhaps $10 billion annually. 

If those figures are anywhere comparable to Android costs, an argument might be made that Android is viable as a stand-alone company.

If an independent Android did not suffer marketplace erosion, and did not incur new costs for access to the rest of the Google ecosystem, one might argue an independent Android could be profitable. 

It is far harder to see how an independent Chrome would fare, as new payment flows would have to be created between Google search and Chrome, for example. And we have at least one good example of a formerly-dominant browser simply disappearing: Netscape Navigator, which was a stand-alone browser with no direct affiliations to a larger entity.

Netscape was dominant between 1994 and 1997. It was crushed by Internet Explorer, which in turn was surpassed by Chrome.

Microsoft Internet Explorer, launched around 1997, took about 90 percent of the market.

Mozilla Firefox launched in 2004 and by 2010, had captured around 30 percent of the market, 

Google Chrome launched in 2008 and became the market share leader, though Microsoft Edge, Safari (Apple), Opera and Brave have some share. 

But Google Chrome remains the dominant browser with over 60 percent global market share across all platforms. Safari holds the second-largest share (approximately 20 percent), largely thanks to Apple’s ecosystem.

The point is that structural separation of some parts of the Alphabet ecosystem, such as Android or Chrome, might not be so smart. The browser market, for example, has seen many changes of market leadership.

Why Firm Productivity Might Drop in the Near Term as AI Gets Deployed

Among other issues, such as potential payback from deploying generative artificial intelligence, is the timing of the payback, and history suggests payback will take far longer than many expect. If AI does develop as a general-purpose technology, as were earlier GPTs including steam power and electricity, and even granting that many technological innovations--which are largely virtual--can propagate much faster than did earlier innovations.  

The initial impact of steam power and electricity on productivity was not as immediate or dramatic as expected. 

Consider steam power. Early adoption was slow. The first practical steam engine was invented by Thomas Newcomen in 1712. James Watt significantly improved the steam engine in 1765 and kicked off the process of commercialization. Still, by 1830, only 165,000 horsepower of steam was in use in Britain, for example.

Even in 1870, about two-thirds of steam power was concentrated in just three industries: coal mining, cotton textiles, and metal manufactures. So, while invented in the early 18th century, it took about 50 to 75 years for steam power to begin having truly widespread and transformative effects on industry and the economy.

The major productivity gains from electricity in the United States came in the 1920s, about 40 years after Thomas Edison first distributed electrical power in New York in 1882.

And there is ample prior evidence of actual productivity dips in  the early days of new technology diffusion. The J Curve, for example, illustrates the pattern that there is an early period of disruption and actual productivity decline when a major new technology is introduced. Only later are the tangible benefits seen. 

source: Flexible Production 

The J-curve effect in GPT adoption typically follows a few stages, from initial investment to realized productivity. AI clearly is in the early investment phase, which ought to imply significant costs without immediate financial returns.

Which ought to clue us in to the fact that investors are likely to be quite disappointed when most entities cannot show significant financial returns. 

And though the J curve might not apply when innovations do not require value chain disruption and displacement, Verizon’s experience with fiber-to-home upgrades still show that even innovations that do not require business model change can take a while to reach maturity. 

As significant as fiber-to-the-home was deemed to be by Verizon, one would be very hard pressed to show significant financial returns to Verizon for five years from mass deployment.

FTTH was not a GPT that required changes in consumer behavior or disruptions of Verizon’s supply and value chains. 

The thing about GPTs (and if AI is a GPT the J curve should apply) is that disruption is required. Still, Verizon arguably reached scale in about four to five years of construction, with very-significant revenue contributions for new video entertainment services enabled by the FiOS network. In the second quarter of 2011, for example, Verizon had about 4.5 million broadband accounts, as well as3.8 million video accounts. 

In the second quarter of  2011, FiOS generated 57 percent of consumer wireline revenues, up from 48 percent a year earlier, Verizon said that year. 

 

By the third quarter of 2011, FiOS accounted for nearly 60 percent of consumer wireline revenues. In the last quarter of 2014, FiOS contributed 75 percent of consumer wireline revenues. Keep in mind that statistic also includes the diminution of Verizon’s landline voice business, plus the maturation and decline of its linear video entertainment business as well. 

In other words, FiOS revenue became the driver of Verizon consumer fixed network revenue in part because the voice and video entertainment businesses declined. 

Year	Cumulative Capital Investment ($B)	Annual FiOS Revenue ($B)	FiOS Subscribers (Millions)
2006	3.6	0.5	0.7
2010	23.0	7.5	4.1
2014	30.0	12.7	6.6
2018	34.0	11.9	6.1
2022	36.5	12.8	6.3

The main take away is that productivity might actually dip in the near term as firms deploy AI technologies.

General-Purpose Technology	Initial Productivity Dip	Adaptation Period	Productivity Surge
Steam Engine	Slow adoption in early 19th century	1820s-1840s	1850s-1890s
Electricity	Limited productivity gains in 1890s-1910s	1920s-1930s	1940s-1950s
Computers	Productivity paradox in 1970s-1980s	1980s-1990s	Late 1990s-2000s
Internet	Initial investment costs in 1990s	Late 1990s-early 2000s	Mid 2000s-present

What "Killer App" Will Emerge from Generative AI?

 Agents are clearly a lead candidate for the artificial intelligence "killer app." Personalization of your digital experience is one thing; anticipation of your needs is something else. 

With the caveat that it is always possible there is no single and universal “killer app” in any computing era, it still is possible that one could emerge for generative artificial intelligence. 

Certainly, key or lead apps have been important in prior waves of computing development. Sometimes  the killer app is clear enough for end users and consumers. At other times it is the business or organization end users or business-to-business use cases that dominate. 

As a rule, B2B value was dominant in the mainframe and minicomputer eras. Since  then, virtually all killer apps can be identified by the consumer apps that surfaced. 

But some innovations, such as app stores or cloud computing, arguably were important as platforms and ways of doing things, rather than specific apps. 

Era/Platform	Killer App(s)	Rationale
Mainframe Era (1960s-1970s)	COBOL.  Batch Processing	Enabled large-scale business applications like payroll, banking, and insurance systems.
Minicomputer Era (1970s-1980s)	VAX/VMS, Accounting Systems	Brought computing power to smaller organizations, particularly in science, manufacturing, and finance.
Personal Computer (PC) Era (1980s-1990s)	VisiCalc (spreadsheet)	The first spreadsheet program, which revolutionized business and financial management.
PC Era (1990s)	Microsoft Office Suite (Word, Excel, etc.)	Dominated office productivity, becoming essential in business, education, and home environments.
PC Era (1990s)	Internet Browsers (Netscape, Internet Explorer)	Opened the gateway to the World Wide Web, fundamentally changing communication and information access.
Web 1.0 Era (late 1990s-2000s)	Email (e.g., AOL, Hotmail)	Email transformed personal and business communication, enabling near-instant global connectivity.
Web 1.0/2.0 Era (early 2000s)	Search Engines (Google)	Google’s search engine made finding information on the web faster and more accurate, changing web usability.
Mobile Era (2000s)	Text and Instant Messaging (WhatsApp and others)	Redefined personal communication with quick, accessible messaging on mobile phones.
Mobile App Era (late 2000s-2010s)	App Stores (Apple App Store, Google Play)	Created an ecosystem where developers could offer mobile apps, enabling smartphone adoption at scale.
Mobile App Era (2010s)	Social Media Apps (Facebook, Instagram)	Changed social interaction, media consumption, and online behavior globally.
Cloud Computing Era (2010s-present)	AWS, Microsoft Azure, Google Cloud	Enabled scalable, on-demand computing infrastructure, transforming how companies build and deploy services.
AI Era (2020s)?	Generative AI (ChatGPT, others)	Revolutionizing content creation, customer service, and automating complex cognitive tasks. Enable AI agents

We don’t yet know what killer apps could emerge in the AI era, but early on, generative AI might be a lead platform. Still, some believe AI agents could emerge as a potential killer app for GenAI.