Google CEO Sundar Pichai has predicted that artificial intelligence could become “bigger than the internet.” But this is a forecast about AI’s potential economic and technological impact—not proof that AI has already surpassed the internet in users, revenue, or social importance.
Pichai made a cautious version of the claim in 2023 and used stronger wording in 2025. His argument is that AI could become a general-purpose platform embedded in search, phones, software, cloud services, businesses, and new products—largely on top of the internet rather than as its replacement.
As an Amazon Associate I earn from qualifying purchases.
What Sundar Pichai actually said
Pichai has made versions of the claim in several settings:
- September 5, 2023: In a Google anniversary essay, he wrote that AI could be “bigger than the internet itself.” He described AI as potentially larger than the shift from desktop computing to mobile.
- February 10, 2025: In remarks at the AI Action Summit, Pichai said AI could democratize access to information more than the internet did.
- May 2025: In an interview discussed by Search Engine Land, he used the more direct wording: “I think AI is going to be bigger than the internet.”
The wording matters. “May be bigger” is a possibility; “is going to be bigger” is a much more confident prediction. Neither statement is an independently measured fact.
#1 Best Overall
His original 2023 essay is available on Google’s website, while his 2025 speech is published in an official Google transcript.
What does “bigger than the internet” mean?
“Bigger” does not have one universally accepted measurement. Pichai’s argument is best understood as a platform-shift argument. He is suggesting that AI could affect more layers of the economy and computing system than the early web did.
1. AI could reach more industries
The internet transformed communication, commerce, advertising, entertainment, software distribution, and media. AI could be used in nearly all of those areas while also changing medicine, education, manufacturing, logistics, finance, customer service, scientific research, and software development.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →2. AI can be embedded everywhere
People may use AI without opening a separate chatbot. It can appear inside search, email, word processors, phones, cameras, cloud applications, customer-service systems, vehicles, and developer tools. That could give AI broad economic reach even when users do not think of themselves as “using AI.”
3. AI changes the computer interface
The web primarily helped people find and exchange information. AI can also generate text, images, code, and audio; summarize documents; translate languages; answer questions; analyze data; and, in some products, use tools or complete multiple steps.
That makes AI potentially more than a destination on the internet. It could become an interface layer through which people access many existing services.
4. AI could create new companies and categories
The internet created search engines, social networks, online marketplaces, streaming services, cloud software, app platforms, and digital advertising businesses. Pichai expects AI to generate an equivalent wave of products and companies, including agent-based software, new hardware, and automated services.
Is AI replacing the internet?
No—not in the literal sense. AI depends heavily on internet connectivity, cloud infrastructure, data centers, software distribution, and online information. Many AI products are accessed through websites or mobile apps and rely on internet-connected data and services.
A more accurate interpretation is that AI could become the next major general-purpose layer built on top of the internet. The internet remains the underlying network, while AI may become a more active way of searching, creating, deciding, and completing tasks.
This distinction matters for investors, businesses, and consumers. A company can benefit from AI growth without AI eliminating the internet businesses that support it. At the same time, AI could shift value away from some existing websites, publishers, software interfaces, and search practices.
Why the prediction matters especially to Google
Google has a direct strategic stake in this transition. The company became one of the internet’s most valuable businesses through Search, advertising, online services, Android, YouTube, and cloud computing.
Free tools Windows power users keep installed
One-click scans. No signup required.
AI creates both an opportunity and a threat:
- Opportunity: Google can provide models, cloud infrastructure, chips, search tools, consumer applications, and enterprise software.
- Threat: Users may ask an AI system for a complete answer instead of visiting Google Search results or clicking through to multiple websites.
- Distribution advantage: Google can place AI into products that already have large user bases, including Search, Android, Gmail, Docs, Photos, and Cloud.
- Competitive pressure: Google must compete with other model developers, cloud companies, device makers, software firms, and startups.
Google’s 2023 anniversary essay presented a progression from Search and mobile computing to machine learning, custom AI chips, the Transformer architecture, and generative AI. Those examples support Google’s strategic case, but they are still company-selected evidence rather than independent proof that AI will exceed the internet’s impact.
What could change for Google Search?
Search is likely to be one of the most important battlegrounds. Google is moving beyond a simple list of links toward AI-generated answers, multimodal queries, AI Mode, and potentially agentic actions.
For users, this could make research faster. Instead of opening several pages, a person may ask a complicated question and receive a synthesized response. AI can also help compare options, interpret images, summarize documents, or guide a multistep task.
There are significant trade-offs:
- Convenience versus verification: A direct answer is faster, but users may be less likely to inspect the underlying sources.
- Answer quality versus source visibility: AI summaries can obscure context, omit important qualifications, or select weak sources.
- Better search versus fewer clicks: If users complete tasks on the results page, publishers and businesses may receive less traffic.
- New interface versus monetization uncertainty: Advertising and referral economics may change when search becomes more conversational and task-oriented.
Google has said it continues to send substantial traffic to the web, but claims about its traffic should be treated as statements from Google rather than as independently settled conclusions. The commercial effect will depend on how often users click sources, how AI answers are monetized, and whether publishers continue producing high-quality information.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteThe next phase: from models to agents
The AI industry’s first phase focused heavily on building larger and more capable foundation models. The next phase is likely to focus on products that use those models to complete workflows.
- Models: Systems that process text, images, audio, video, and code.
- Products: Applications that use models for writing, research, programming, design, customer support, or data analysis.
- Agents: Systems that can plan, browse, call tools, operate software, and complete multistep tasks within defined permissions.
- New hardware: Potential interfaces involving wearables, cameras, vehicles, robotics, and ambient assistants.
These systems could create significant productivity gains, but advertised capability is not the same as reliable autonomy. Users still need to consider permissions, human review, data access, error handling, and the cost of mistakes.
The strongest case for Pichai’s prediction
There is a credible argument that AI could eventually have an impact comparable to or greater than the internet:
- It is general-purpose and can be applied across nearly every sector.
- It can be integrated into existing software rather than requiring users to learn an entirely new destination.
- It supports natural-language, visual, and voice interfaces.
- It may lower the cost of software development and specialized analysis.
- It can augment research, education, creative work, and knowledge-intensive jobs.
- It could create new businesses while changing how existing companies operate.
Google highlights examples involving multimodal Gemini systems, Workspace, Search, Cloud, research, translation, accessibility, and climate-related applications. Google also argues that improvements in efficiency and falling processing costs can make AI more accessible to developers. These are reasonable indicators of investment and experimentation, but they do not yet establish the long-term economic outcome.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Why investors and consumers should remain skeptical
Pichai’s claim is an executive forecast and should be evaluated against measurable results rather than accepted as a conclusion.
AI depends on the infrastructure it supposedly surpasses
AI models require networks, data centers, electricity, chips, cloud platforms, and online data. It may be more accurate to say that AI will become a powerful new use of the internet than to describe it as an independent successor.
Reliability is unresolved
AI systems can produce incorrect answers, invent citations, misunderstand instructions, or take an inappropriate action. These problems are especially important in finance, medicine, law, education, cybersecurity, and business operations.
Deployment is expensive
Building and operating advanced AI requires specialized chips, energy, data centers, engineers, security systems, and evaluation processes. Lower model prices could increase adoption, but total business costs also include integration, governance, monitoring, and employee training.
Adoption may be uneven
Usage varies by country, language, income, device, account type, regulation, and access to fast internet. AI could widen inequality if its benefits primarily reach well-funded companies and highly connected populations.
Economic disruption could be substantial
AI may automate some tasks while increasing demand for other skills. The result could include job displacement, wage pressure in selected occupations, new forms of work, and greater returns for companies that control models, data, chips, or distribution.
Power could become more concentrated
Frontier models and computing infrastructure are expensive. If a small number of companies control the most capable systems and distribution channels, competition, privacy, and bargaining power could become major concerns.
Risks beyond the technology
The social and political risks include:
- Job displacement and labor-market disruption.
- Copyright and training-data disputes.
- Privacy loss and expanded surveillance.
- Disinformation and synthetic media.
- Bias or unequal performance across languages and populations.
- Cybersecurity misuse.
- Energy and water consumption from data-center expansion.
- Overreliance on incorrect or overly confident answers.
- An “AI divide” between people and regions with unequal access to computing, education, and connectivity.
Pichai’s official remarks acknowledge that AI creates risks and require responsible development, policy coordination, security, and broad access. Those are Google’s stated priorities, not evidence that the risks have been solved.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallWhat the prediction means for ordinary users
Consumers are likely to encounter AI through products they already use rather than through one single application. Examples include:
Best Value
- AI-generated or AI-assisted search answers.
- Writing, meeting, spreadsheet, and email tools.
- Image editing and generation.
- Research assistants that work with personal documents.
- AI features built into smartphones and operating systems.
- Developer tools that generate or explain code.
Google’s consumer AI offerings can include Gemini access, storage, and integrations with Google services. Availability, limits, model names, and features vary by country, language, device, account, and subscription. A subscription also does not guarantee accuracy or professional-grade performance.
What it means for businesses and developers
Businesses should separate several very different types of Google AI product:
- Consumer subscriptions: Flat-rate plans aimed at individuals and bundled with Google services.
- Workspace AI: Features and licensing for organizations using Gmail, Docs, Sheets, Meet, and Drive.
- Cloud and developer products: Usage-based model access, enterprise controls, infrastructure, and coding tools.
- Custom integrations: Systems that require data governance, security reviews, testing, monitoring, and employee training.
Google Cloud generally bills generative AI usage according to factors such as model, input and output tokens, context, modality, batch processing, and grounding. Current rates should be checked on the official Vertex AI pricing page, because model names and prices change.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Workspace billing also differs between flexible and annual or fixed-term commitments. Organizations should consult Google’s official billing guidance rather than relying on an old price table.
How to judge whether AI really becomes “bigger”
The prediction can eventually be tested using several measures:
- Reach: How many people and organizations use AI regularly?
- Economic value: Does AI generate durable revenue, productivity gains, and new businesses?
- Infrastructure: Does AI become a foundational layer across computing?
- Behavior: Do people change how they search, work, learn, create, and communicate?
- New categories: Does AI produce important products that could not otherwise exist?
- Durability: Does adoption persist after promotional subsidies and novelty fade?
- Distribution: Do the benefits reach lower-income regions and non-English speakers?
- Reliability: Can people safely delegate meaningful tasks?
- Concentration: Are benefits widely distributed or controlled by a small group of firms?
Bottom line: a powerful forecast, not a proven fact
Sundar Pichai is not saying that AI will literally eliminate the internet. He is arguing that AI could become a more consequential general-purpose platform—one that changes search, software, hardware, business operations, productivity, and the way people interact with information.
That is a defensible description of AI’s potential, but it remains a prediction. The meaningful test will be whether AI produces lasting changes in everyday behavior, business formation, productivity, infrastructure, and the distribution of economic power over the next decade. For consumers and investors, the sensible response is to track measurable adoption and results rather than treating an executive’s headline-making forecast as an established fact.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
