What is disruptive innovation curve?
In 1995, Harvard Business School professor late. Clayton Christensen proposed the theory of Disruptive innovation as a process where a product or service typically exhibits below behaviour:
- Initial Low Performance: Disruptive innovations start with lower performance but target niche or underserved segments.
- Meeting Customer Needs: As the innovation improves, it begins to meet the performance levels required by mainstream customers.
- Overtaking Incumbents: Eventually, the disruptive technology surpasses incumbents, often rendering the old technology obsolete.
What did the Disruptive Innovation Curve look like during the PC era?
| Phase | PC Performance | Customer Needs | Incumbents Focus |
| Early 1970s – 1980s | PCs were underpowered compared to mainframes & minicomputers | PCs were marketed to hobbyists, small businesses and consumers | Focussed on high even businesses & ignored personal computers |
| Growth 1980s – 1990s | PCs became more powerful using Intel 386 & 486 CPUs. Windows & Macs provided GUI & built business apps like spreadsheets | PCs became good enough for business use, students. | Falling costs of CPUs, Memory, Storage made PCs affordable becoming mass market |
| Disruption 1990s – 2000s | PCs became powerful enough to handle tasks traditionally powered by mainframes & minicomputers. | As Internet became widely available, PCs became essential & were adopted by families | IBM which was a dominant force in mainframes, eventually had to exit the PC market |
The graph illustrates the disruptive innovation curve for personal computers (PCs) from 1970 to 2000. It highlights how PCs started as a low-performance alternative to mainframes and minicomputers, targeting niche underserved markets like for hobbyists, but rapidly improved over time to meet and exceed mainstream customer needs, eventually overtaking the incumbents in capability and market share.
What did the disruptive innovation curve look like for cloud software?
| Phase | Cloud Performance | Customer Needs | Incumbents Focus |
| Early 2000 – 2010 | Early cloud solutions like Salesforce & AWS had limited functionality compared to on-premise software | Cloud companies targeted startups that didn’t have resources to invest in capex & liked the pay-per-use model | On premise software catered to large corporations with mission critical apps |
| Growth Mid 2010s | With broadband Internet, more demanding applications were being hosted on the cloud | Mid & large size businesses started to experiment with non-critical applications like CRM etc | Incumbents still focussed on large enterprise applications with huge capex |
| Disruption Late 2010s – Present | With high speed Internet, cloud software became as good as on premise | Enterprises also migrated critical apps like ERP to cloud providers like AWS, Microsoft Azure & Google Cloud | Incumbents like Microsoft pivoted to offering cloud hosted apps while many others perished |
This curve showcases how cloud software disrupted traditional on-premises software by starting at the low end and rapidly climbing to market dominance.
What did the disruptive innovation curve look like for smartphones?
| Phase | Smartphone Performance | Customer Needs | Incumbents Focus |
| Early 1990s – mid 2000s | Blackberry & Palm Pilot had poor battery life & signal reception compared to Nokia feature phones | Focussed on an underserved niche, business users in this case who valued mobile email, calendar & contacts. | Incumbents focussed on mass market feature phones with 2 day battery life & more affordable. |
| Growth 2007 – 2011 | iPhone & Android smartphones with touchscreens & mobile internet began to redefine what smartphones can do | Good enough to meet daily needs of mass markets such as web browsing, casual gaming, social media, chat & media consumption | Initially Nokia & Windows Phone fought by upgrading their own OS and devices but couldn’t keep up with the innovation |
| Disruption 2011 – 2015 | iPhone & Android outperformed feature phones in battery, storage, camera, apps availability and every other parameter | With app store & play store, iPhone & Android had millions of applications that consumers & business users needed | Heavily disrupted, Nokia & Windows Phone exited the business |
What would the disruptive innovation curve look like for AI?
| Phase | AI Performance | Customer Needs | Incumbents Focus |
| Early 2024 – 2025 | Niche use cases such as generative text, image & code as well as data analysis. | Customer needs require nuanced understanding, long-term memory, and complex decision-making which AI currently doesn’t support | Incumbent effort is human led |
| Growth 2025 – 2030 | AI will tackle more common use-cases such as Search, Shopping, Workflow management across industries. | Will become cheaper than current $20/mo or even free by introducing ad-supported tier for mass market adoption. Apps & services will be AI native and AI first. | Incumbents like Google & Amazon will be forced to pivot into an AI first business model |
| Disruption 2030 – 2040 | AGI would be a reality | AGI reaching human level intelligence, AI personal assistants or mentor may be a reality. | Given insane capex, consolidation is inevitable. |
In many industries, their leaders and employees are ignoring the disruptive innovation curve at their own peril because they don’t yet see parallels to the past disruption events. They are dismissive of AI because it cannot perform complex tasks with multi-step workflows and past context or make decisions at human intelligence levels (yet). They miss the keyword “yet”.
What to expect during the AI growth phase between 2025 – 2030?
Between 2025 and 2030, we can expect AI to transition from niche applications to a diverse range of mainstream use cases. AI will no longer be limited to assisting writers, developers, and designers but will become an integral part of everyday consumer experiences.
One of the most significant disruptions on the horizon is search—a domain long dominated by Google. Since the early 2000s, Google has perfected its ad-centric business model, innovating to deliver highly targeted text, display, video, and product ads across devices. This approach drove exponential growth in Google’s revenue, profitability, and stock price. However, the user experience often suffered. Content farms manipulated Google’s algorithms, flooding search results with clickbait articles, forcing users to sift through countless ads and filler words to find meaningful information.
Interestingly, Google’s own researchers invented the Transformer architecture in 2017, a breakthrough technology that could have improved search experiences by providing users with precise answers. Yet, leveraging this technology risked disrupting their ad-driven business model—a classic case of the innovator’s dilemma. In contrast, OpenAI embraced this architecture, launching ChatGPT in November 2022. OpenAI and similar companies like Perplexity are now reshaping search by introducing AI-powered “answer engines” that summarize relevant information, saving users from the frustration of ad-infested web pages.
Beyond search, AI is poised to revolutionize other industries. In travel, AI-first platforms could offer fully personalized booking experiences—identifying the best destinations, accommodations, and itineraries based on individual preferences. This goes beyond the search-and-book functionality of traditional online travel agencies.
Similarly, AI could redefine online shopping. Imagine an AI-powered fashion designer creating personalized wardrobes by considering body shape, climate, style preferences, and budget. Or an AI interior designer analyzing photos and videos of a room to recommend decor and furniture that fits the space, accounts for local trends, and factors in material availability.
As these mainstream use cases proliferate, AI will no longer be a luxury or novelty—it will be a foundational part of how consumers interact with technology, shaping the next wave of innovation.
What to expect during the AI disruption phase 2030 – 2040?
Artificial General Intelligence (AGI) is on the horizon, poised to transform the very fabric of human potential. Unlike narrow AI, which excels in specific tasks like language translation or image recognition, AGI will possess the ability to perform any intellectual or complex task a human can, with comparable understanding, adaptability, and learning capacity. AGI’s generalization capabilities will enable it to apply its knowledge and skills across a vast range of domains and situations, making it a revolutionary leap in artificial intelligence.
With AGI, the concept of AI personal mentors, assistants, and coaches could become a reality. An AI mentor would excel in problem-solving, logical reasoning, creativity, planning, decision-making, and even emotional understanding. Imagine a mentor capable of guiding you through unfamiliar challenges by leveraging insights from diverse fields. It could help you set and achieve meaningful goals by deeply understanding your strengths, weaknesses, context, and objectives.
Such an AI coach or mentor could act as a digital twin, representing you in situations where your physical presence is impossible, or providing actionable advice tailored precisely to your needs. From strategic career guidance to enhancing personal growth, an AI mentor would serve as an indispensable ally, amplifying human capabilities and redefining the boundaries of achievement.
Naysayers will continue to compare AI’s performance against humans until it becomes good enough for mainstream use cases. But it’s an inescapable eventuality.