As artificial intelligence rapidly transforms from research curiosity to essential technology, questions about its future impact, economic implications, and global distribution of benefits become increasingly important. In this final part of our series with Aravind Srinivas, CEO of Perplexity AI, we explore predictions for AI’s evolution over the next five years, examine opportunities for entrepreneurs (particularly in emerging markets like India), and consider appropriate regulatory approaches. Understanding these forward-looking perspectives provides valuable context for navigating a world increasingly shaped by AI capabilities.
Main Body
The Next Five Years: From Assistants to Personalized Agents
When asked to predict AI’s evolution over the next five years, Aravind envisions a world where personal AI assistants become ubiquitous and democratized: “We’ll all have a personal assistant. It’s going to feel really amazing. It’s not going to be a luxury thing anymore.”
He draws an analogy to the iPhone’s democratizing effect on technology: “The same phone that the president of the US uses, you’re going to be able to use too.” This accessibility will fundamentally change how people interact with technology and accomplish tasks.
Aravind predicts two transformative shifts:
- Ubiquitous Personal Assistance: AI assistants will handle everything from scheduling to communication, operating independently while their human users focus elsewhere: “I’m just in my bed sleeping and the AI is working for me.”
- Personalized Creation: “People are going to be able to build personalized things for them…whatever you want to exist in the world, you can make it happen.” This represents a shift from consuming others’ creations to easily building custom solutions.
However, this optimistic vision comes with challenges: “The dystopian part is, unfortunately, in the short term, there’s going to be a lot of labor displacement. Not as many people are needed to get work done anymore.” This creates economic questions about how people will adapt and find meaningful work.
Aravind notes that the current trend of building smaller companies with outsized impact will intensify: “You don’t need 10,000 people companies to be a trillion-dollar company anymore.” This efficiency creates both opportunities and disruptions in the labor market.
Entrepreneurial Opportunities in the AI Era
For entrepreneurs looking to build in this new landscape, Aravind identifies several promising directions:
recommends a staged approach: “Build a product that’s pretty interesting and new, get users, raise more money, get more users…start with post-training on top of open-source models, then start to look into pre-training too, and then get into the data centers.”
Appropriate Regulation for AI
On the question of how governments should regulate AI, Aravind advocates a light touch focused on specific applications rather than fundamental models: “Regulating models is not necessarily a great idea, and it’s not going to work in practice either.”
He identifies one area warranting attention: “What I find pretty concerning is probably people using chatbots when they’re kids and developing relationships with them…and feeling suicidal when they don’t get to enjoy the chatbots anymore.” This represents a new kind of psychological risk that might justify targeted regulation.
However, he cautions against broader restrictions: “We’re kind of still very early in AI. Moving slow is going to cost us a lot long-term…hundreds of billions or trillions of dollars.” Instead, he recommends “keep accelerating right now and be mindful of use cases…that are clearly dangerous, but otherwise be pretty open-minded and build stuff and see how things play out.”
Conclusion
The AI revolution is accelerating, transforming from systems that answer questions to agents that take action on our behalf. This shift promises to democratize capabilities once reserved for the privileged few, enabling everyone to create personalized solutions and accomplish more with less effort.
For entrepreneurs, particularly those in emerging markets, opportunities exist in developing specialized AI applications for underserved populations, creating platforms that enable personalization at scale, and building infrastructure that supports sovereignty and security requirements.
As we navigate this transition, balancing innovation with appropriate safeguards will be crucial. Rather than attempting to regulate fundamental technology, focusing on specific harmful applications allows us to capture AI’s benefits while mitigating its risks.
The future Aravind envisions is neither utopian nor dystopian but a complex mixture of new possibilities and challenges. Those who understand both the technology and its human implications will be best positioned to shape this future in positive directions.