10 Themes for 2026

AI scaling era has begun

Jan 06, 2026

Hello 2026! Sharing a few themes that are top of mind heading into the year.

The vibe heading into 2026 feels different, a more sober tone. Less exuberance and more of a “build, deploy, scale, execute on all fronts” type energy.

Early AI winners are accelerating + compounding, scale of everything is unprecedented. It’s the era of $100B startups. Constant mental resets required.
Demand is insatiable with customers all in market at once but not forever. In 2026 enterprise buyers are more discerning; consumers are cementing habits.
Plenty of emerging categories still up for grabs but in every space there is intense pressure for startups to break away from the pack. Velocity beats predictability.
Talent is once again the rate-limiting factor to growth, even for proudly lean startups. Building the human system that builds the company is still hard.
The friction to AI adoption is very human right now. Gaps in awareness, usability, trust, skill, fear of job loss and individual ability to absorb change.
With great power comes great responsibility. Societal skepticism of AI is something every Founder / CEO will need to contend with.

The AI scaling era is here. The stakes are high. Lock in. That’s the vibe.

Top 10 themes for 2026

1. Asynchronous, always-on agents take off. Agents are already remarkably good at searching, reasoning, performing discrete tasks, coding, especially when given context and constraints. We can now offload all manner of known, ongoing, atomized work and do it 24x7 in the background (for example setting Tasks on Perplexity, Scouts on Yutori, enterprise data workloads on Parallel, coding in Claude). There is already an invisible agentic workforce at our disposal and we’re learning how to assemble and manage it. For startups, mastering the async “set it and forget it” mode is a sneaky good way to create base-load demand and embed in repeat workflows, which is incredibly sticky. Great for retention and recurring revenue.

2. Context is king and personalization is how we make it useful. That realization we have with our Instagram feeds — it really knows us! Well we’re not there yet for AI tools but we are getting to a remarkable level of personalization-enabled utility with context, memory and more months of collecting user data. I’m feeling it in Perplexity, for example. It knows what I do, what type of work product I like, what domains I spend a lot of time in. It combines context from previous queries and multiple aspects of my life when providing the best possible recommendation. It’s hard to imagine switching and starting with a blank slate in another product. Aggregating context is more powerful than simply aggregating attention.

3. Consumer renaissance is flourishing and creating new categories. VCs love to say there’s nothing happening in consumer but if you aren’t seeing it, maybe you’re not paying attention. Where founders are finding success in building new consumer applications and getting to escape velocity is 1) making the luxury affordable and accessible (AI therapist, health coach, wealth advisor, SAT prep, etc), 2) giving naturally creative people new mediums to self-express (music, video, voice), 3) enabling anyone to be a software creator (Lovable, Wabi). Good ‘ole network effects, multi-player, brand building, social proof and cultural relevance still work because people buy products they love, find useful and their friends use. It’s not all about AI.

4. AI-native wearables, the trojan horse for context farming. Many hardware startups have tried and failed but the “why now” for AI wearables is clearer in 2026 than before. New devices are predicated on ambient recording, which seemed invasive and socially awkward two years ago but now seems normal and even useful (e.g. Granola). That’s net new behavior and it opens the door for an AI device that collects context on the go. Will users trust OpenAI with even more of their data? Will Meta’s talent binge pay off in some awesome new AI products? How does Apple respond? Remember, consumers love free and have historically been willing to trade privacy for utility (e.g. TikTok, Google). Gen Z puts it this way, “they already have all of my data, what’s a little more.” So the interesting question isn’t will it happen but what is the killer use case? Who gets there first? Who wins the trust and the zeitgeist?

5. Agentic Commerce. AI search won, consumers voted with their attention and they’re not going back. It reminds me of Amazon 1-click buy back in the day. Once you experience it, you don’t want to shop without it and you expect it everywhere. The interesting questions in agentic commerce are 1) what is the source of truth for user-agents fetching and reasoning through product options, 2) what is authoritative in agent-driven discovery and decisions, 3) how do brands and retailers merchandise and market through agents to consumers and 4) what is the personalized native ads format? As in prior commerce shifts, the battle ground isn’t for checkout per se, it’s who owns first touch (the decision funnel has collapsed) and who owns the consumer data.

6. Enterprise adoption accelerates: pickier buyers, bigger deployments. This year we’ll see a productive convergence of enterprise decision-makers becoming much more informed and intentional about their AI strategy and AI-native startups having had the time to mature and get products enterprise-ready. Enterprise adoption requires solving for eng reliability, security, accuracy, auditability, permissioning, etc. AI-native startups are finally getting there. Expect enterprise buyers to be more selective and favor startups that sell a whole solution, not a tool (FDE FTW) but also more prepared to deploy AI products at scale and to more of their workforce.

7. Functional agents for company operations. Enterprises are complex ecosystems with many supporting functions to manage day to day operations. There is a thick services layer within large companies and often outsourced labor too. Supporting functions are incredibly important and at the same time there is sooo much wasted effort in managing across silos, systems and complex internal processes. At best, it’s grunt work. At worst, it’s drag on the business creating information asymmetry, decision latency, operating overhead. Now we can enable Finance, Compliance, Marketing, Legal, Procurement, RevOps, IT / HR with composable workflow agents and intelligent systems of action. This is where we’ll see exciting AI App winners. The interesting thing is how does AI change the org chat over time? What new functions emerge? Which ones collapse together or go away entirely?

8. Voice agents and conversational AI. This market is taking off and the opportunity is massive, we’re just scratching the surface. I had a medical issue recently and spent way too long waiting on the phone to schedule an appointment and the only way to do it was during regular business hours. Voice agents can service customers 24x7 and in many ways perform better than a cranky, distracted human, ideally with more context. The ubiquitous deployment of intelligent voice agents for everything (customer front-end, customer support, intakes, outbound, etc) can’t come soon enough.

9. Self-learning AI models and systems. This is one of the well-covered near frontiers and it’s interesting to think about how we can move from generalist models to last-mile optimization for specific domains to self-learning paradigms, all in service of attaining significantly “better than expert human” performance. I’ll let AI researchers debate the merits of their various approaches but at a high level there’s more teams chasing this than the big AI labs, one of them might yet surprise us.

10. The $1T IPO. Historic significance if it happens. Fundamentally resets expected outcomes in venture, founder aspirations and public perception of tech. In the meantime, mega deals will get done. Looming elections create time pressure to transact now. We’ll see more M&A deals and license-hires from tech giants and vulnerable enterprise software companies. Prepare for 2026 to feel a bit different. More of the tone set by deals and exits vs. awe-inspiring, bottoms-up innovation.

If you made it this far… thank you. You’re a champ.

What themes are top of mind for you in 2026?

Shortcuts by Ann Bordetsky

Discussion about this post

Ready for more?