Written by: a16z New Media

Translated by: Block unicorn

Yesterday, we shared the first part of the "Major Concepts" series, which includes insights from our infrastructure, growth, bio + health, and the Speedrun team partners on the challenges startups will face in 2026.

Today, we continue to roll out the second part of the series, featuring contributions from American Dynamism (a16z's investment team established in 2021) and the applications team.

American Dynamism

David Ulevitch: Building an AI-Native Industrial Foundation

America is rebuilding the economic components that truly empower the nation. Energy, manufacturing, logistics, and infrastructure are back in focus, with the most significant shift being the rise of a genuinely AI-native, software-first industrial foundation. These companies start with simulation, automated design, and AI-driven operations. They are not modernizing the past; they are building the future.

This is creating tremendous opportunities in advanced energy systems, heavy robotics manufacturing, next-generation mining, and biological and enzymatic processes (producing precursor chemicals relied upon by various industries). AI can design cleaner reactors, optimize extraction, create better enzymes, and coordinate autonomous machine clusters with insights unattainable by traditional operators.

The same transformation is reshaping the world beyond factories. Autonomous sensors, drones, and modern AI models can now continuously monitor critical systems that were once large-scale and difficult to manage, such as ports, railways, power lines, pipelines, military bases, and data centers.

The real world needs new software. Founders building this software will shape America's prosperity in the next century.

Erin Price-Wright: The Revival of American Manufacturing

America's first great century was built on strong industrial power, but as we know, we have lost much of that industrial strength—partly due to offshoring and partly due to a deliberate societal lack of constructive building. However, rusty machines are coming back to life, and we are witnessing a revival of American manufacturing centered around software and AI.

I believe that by 2026, we will see companies responding to challenges in energy, mining, construction, and manufacturing with a factory mindset. This means combining AI and automation technologies with skilled workers to make complex, customized processes as efficient as assembly lines. Specifically, this includes:

• Rapidly and repeatedly responding to complex regulations and permitting processes
• Accelerating design cycles and conducting manufacturability design from the outset
• Better managing large-scale project coordination
• Deploying autonomous systems to accelerate tasks that are difficult or dangerous for humans

By applying the technologies developed by Henry Ford a century ago, planning for scale and repeatability from the start, and integrating the latest advancements in AI, we will soon achieve mass production of nuclear reactors, build housing to meet national demand, construct data centers at astonishing speeds, and enter a new golden age of industrial strength. As Elon Musk said, "Factories are products."

Zabie Elmgren: The Next Wave of Observability Will Be Physical, Not Digital

Over the past decade, software observability has changed how we monitor digital systems, making codebases and servers transparent through logs, metrics, and tracing. The same transformation is about to sweep through the physical world.

As major cities in the U.S. deploy over a billion connected cameras and sensors, physical observability—real-time awareness of the operational status of cities, power grids, and other infrastructure—is becoming both urgent and feasible. This new layer of perception will also drive the next frontier of robotics and autonomous technologies, where machines will rely on a universal framework to make the physical world as observable as code.

Of course, this shift also carries real risks: tools that can detect wildfires or prevent construction site accidents could also trigger dystopian nightmares. The winners of the next wave will be those who earn public trust and build privacy-protecting, interoperable systems that natively support AI, thereby enhancing societal transparency without compromising social freedoms. Those who can construct such a trustworthy framework will define the direction of observability in the next decade.

Ryan McEntush: The Electronic Industrial Stack Will Change the World

The next industrial revolution will not only occur in factories but also within the machines that power them.

Software has fundamentally changed how we think, design, and communicate. Now, it is changing how we travel, build, and produce. Advances in electrification, materials, and AI are converging to enable software to truly control the physical world. Machines are beginning to perceive, learn, and act autonomously.

This is the rise of the electronic industrial stack—a comprehensive technology powering electric vehicles, drones, data centers, and modern manufacturing. It will connect the atoms of the world with the bits that control them: from minerals refined into components, energy stored in batteries, electricity controlled by electronic devices, to motion achieved through precision motors, all coordinated by software. It is the invisible foundation behind every breakthrough in physical automation; it determines whether software merely summons a taxi or truly takes the wheel.

However, the ability to build this stack is dwindling, from refining key materials to manufacturing advanced chips. If America wants to lead the next industrial era, it must manufacture the hardware that supports it. Nations that master the electronic industrial stack will define the future of industrial and military technology.

Software has consumed the world. Now, it will drive the world forward.

Oliver Hsu: Autonomous Laboratories Accelerate Scientific Discovery

With advancements in multimodal model capabilities and continuous improvements in robotic operational abilities, teams will accelerate autonomous scientific discovery. These parallel technologies will give rise to autonomous laboratories capable of closing the loop on scientific discovery—from hypothesis generation to experimental design and execution, to reasoning, result analysis, and iteration on future research directions. Teams building these laboratories will be interdisciplinary and will integrate expertise from AI, robotics, physics and life sciences, manufacturing, operations, and more, enabling continuous experimentation and discovery across fields through unattended laboratories.

Will Bitsky: The Data Journey of Key Industries

In 2025, the zeitgeist of AI will be defined by the limitations of computing resources and data center construction. By 2026, it will be defined by the limitations of data resources and the next frontier of the data journey—our key industries.

Our key industries remain a treasure trove of potential, unstructured data. Every truck departure, every meter reading, every maintenance task, every production run, every assembly, and every test run provides material for model training. However, terms like data collection, labeling, or model training are not commonly used in the industrial sector.

The demand for this type of data is relentless. Companies like Scale, Mercor, and AI research labs are tirelessly collecting process data (not just "what was done," but "how it was done"). They pay a premium for every piece of "sweatshop data."

Industrial companies with existing physical infrastructure and labor forces have a comparative advantage in data collection and will begin to leverage this advantage. Their operations will generate massive amounts of data that can be captured at nearly zero marginal cost and used to train their own models or licensed to third parties.

We should also expect a surge of startups to emerge and provide assistance. Startups will offer a coordinated stack: software tools for data collection, labeling, and licensing; sensor hardware and software development kits (SDKs); reinforcement learning (RL) environments and training pipelines; and ultimately, their own intelligent machines.

Applications Team

David Haber: AI Reinforces Business Models

The best AI startups are not just automating tasks; they are amplifying the economic benefits for customers. For example, in contingency fee-based law, law firms only earn when they win. Companies like Eve leverage proprietary outcome data to predict case success rates, helping firms choose more suitable cases, serve more clients, and improve win rates.

AI itself can reinforce business models. It not only reduces costs but also generates more revenue. By 2026, we will see this logic extend across industries as AI systems align more deeply with customer incentive structures, creating compound advantages unattainable by traditional software.

Anish Acharya: ChatGPT Will Become the AI App Store

The consumer product cycle requires three elements to succeed: new technology, new consumer behavior, and new distribution channels.

Until recently, the AI wave met the first two conditions but lacked new native distribution channels. Most products grew relying on existing networks like X or word-of-mouth.

However, with the release of the OpenAI Apps SDK, Apple's support for mini-programs, and the introduction of group chat features in ChatGPT, consumer developers can now directly tap into ChatGPT's 900 million user base and leverage new mini-program networks like Wabi for growth. As the final link in the consumer product lifecycle, this new distribution channel is expected to spark a once-in-a-decade gold rush in consumer technology by 2026. Ignoring it will have consequences.

Olivia Moore: Voice Agents Are Starting to Take Their Place

In the past 18 months, the vision of AI agents handling real interactions for businesses has shifted from science fiction to reality. Thousands of companies, from small to large, are using voice AI to schedule appointments, complete bookings, conduct surveys, gather customer information, and more. These agents not only save costs and generate additional revenue for businesses but also free up employees to engage in more valuable—and enjoyable—work.

However, as this field is still in its infancy, many companies remain in the "voice as a point of entry" phase, offering only one or a few types of calls as a standalone solution. I am excited to see voice assistants expand to handle entire workflows (potentially multimodal) and even manage complete customer relationship cycles.

This likely means that agents will be more deeply integrated into business systems and given the freedom to handle more complex types of interactions. As underlying models continue to improve—now agents can invoke tools and operate across different systems—every company should deploy voice-first AI products and leverage them to optimize critical aspects of their business.

Marc Andrusko: Proactive Applications Without Prompts Are Coming

By 2026, mainstream users will say goodbye to prompt boxes. The next generation of AI applications will not display prompts at all—they will observe your actions and proactively offer suggestions for your operations. Your integrated development environment (IDE) will suggest refactoring before you even ask. Your customer relationship management (CRM) system will automatically generate follow-up emails after you finish a call. Your design tools will generate various options while you work. Chat interfaces will merely be auxiliary tools. Today, AI will become the invisible scaffolding that runs through every workflow, activated by user intent rather than commands.

Angela Strange: AI Will Ultimately Upgrade Banking and Insurance Infrastructure

Many banks and insurance companies have integrated AI features such as document import and AI voice agents into their traditional systems, but only by rebuilding the infrastructure that supports AI can it truly transform the financial services industry.

By 2026, the risks of failing to modernize and fully leverage AI will outweigh the risks of failure itself, at which point we will see large financial institutions abandon contracts with traditional vendors in favor of newer, more AI-native alternatives. These companies will break free from the constraints of past classifications, becoming platforms capable of centralizing, standardizing, and enriching underlying data from traditional systems and external sources.

What will the results be?

• Workflows will be significantly streamlined and parallelized. No longer will there be a need to switch back and forth between different systems and screens. Imagine being able to see and process hundreds of pending tasks in a loan origination system (LOS) all at once, with agents even able to handle some of the more tedious parts.
• The classifications we are familiar with will merge to form larger categories. For example, customer KYC, account opening, and transaction monitoring data can now be unified within a single risk platform.
• The winners of these new classifications will be ten times the size of legacy enterprises: the scope of classification has expanded, while the software market is consuming the workforce.

The future of financial services is not about applying AI on top of old systems, but about building an entirely new operating system based on AI.

Joe Schmidt: Forward-Looking Strategies Bring AI to 99% of Businesses

AI is one of the most exciting technological breakthroughs of our lifetime. However, so far, most of the gains from new startups have flowed to the 1% of companies in Silicon Valley—either those truly located in the Bay Area or part of its vast network. This is understandable: entrepreneurs want to sell products to companies they are familiar with and can easily access, whether by visiting their offices in person or connecting through venture capitalists on their boards.

By 2026, this situation will change dramatically. Businesses will realize that the vast majority of AI opportunities exist outside of Silicon Valley, and we will see new startups leveraging forward-looking strategies to uncover hidden opportunities within large traditional vertical industries. There are enormous opportunities in traditional consulting and service industries (such as system integrators and implementation companies) as well as in slower-growing sectors like manufacturing.

Seema Amble: AI Creates New Coordination Layers and Roles in Fortune 500 Companies

By 2026, companies will further shift from isolated AI tools to multi-agent systems that need to operate like coordinated digital teams. As agents begin to manage complex and interdependent workflows (such as joint planning, analysis, and execution), businesses will need to rethink the structure of work and how context flows between systems. We are already seeing companies like AskLio and HappyRobot undergoing this transformation, deploying agents throughout processes rather than in single tasks.

Fortune 500 companies will feel this shift most profoundly: they hold the largest reserves of siloed data, institutional knowledge, and operational complexity, much of which resides in employees' minds. Transforming this information into a shared foundation for autonomous workers will unleash faster decision-making, shorter cycles, and end-to-end processes that no longer rely on continuous human micromanagement.

This shift will also force leaders to rethink roles and software. New functions will emerge, such as AI workflow designers, agent supervisors, and governance leads responsible for coordinating and auditing collaborative digital workers. In addition to existing record systems, companies will need coordination systems: new layers to manage multi-agent interactions, assess context, and ensure the reliability of autonomous workflows. Humans will focus on handling edge cases and the most complex situations. The rise of multi-agent systems is not just another step in the automation process; it represents a reconstruction of how businesses operate, make decisions, and ultimately create value.

Bryan Kim: Consumer AI Shifts from "Help Me" to "Understand Me"

2026 will mark a shift in the functionality of mainstream consumer AI products from enhancing productivity to enhancing interpersonal connections. AI will no longer just help you get work done; it will help you understand yourself more clearly and build stronger relationships.

It is important to clarify: this is no easy task. Many social AI products have been launched but ultimately failed. However, thanks to multimodal context windows and continuously decreasing reasoning costs, AI products can now learn from all aspects of your life, not just what you tell the chatbot. Imagine your phone's photo album showcasing genuine emotional moments, one-on-one messages and group chat modes changing based on the conversation partner, and your daily habits shifting under stress.

Once these products truly hit the market, they will become part of our daily lives. Generally speaking, "understand me" products have better user retention mechanisms than "help me" products. "Help me" products profit from users' high willingness to pay for specific tasks and focus on improving user retention. "Pay attention to me" products profit from ongoing daily interactions: users have a lower willingness to pay, but retention rates are higher.

People have been continuously exchanging data for value: the question is whether the returns they receive are worth it. The answer will soon be revealed.

Kimberly Tan: New Model Primitives Give Rise to Unprecedented Companies

By 2026, we will witness the rise of companies that could not have existed before breakthroughs in reasoning, multimodality, and computer applications. So far, many industries (such as legal or customer service) have leveraged improved reasoning technologies to enhance existing products. But we are just beginning to see some companies whose core product functionalities fundamentally rely on these new model primitives.

Advancements in reasoning capabilities can give rise to new abilities for assessing complex financial claims or taking action based on dense academic or analyst research (for example, adjudicating billing disputes). Multimodal models make it possible to extract potential video data from the physical world (for example, cameras on manufacturing floors). The application of computers enables the automation of large industries that have historically been constrained by desktop software, poor APIs, and fragmented workflows.

James daCosta: AI Startups Scale by Selling Products to Other AI Startups

We are in the midst of an unprecedented wave of company creation, primarily driven by the current AI product cycle. But unlike previous product cycles, existing enterprises are not sitting idly by; they are also actively adopting AI. So how can startups win?

One of the most effective and underrated ways for startups to outpace existing enterprises in distribution channels is to serve them from the very beginning: that is, to serve those newly established greenfield companies (i.e., entirely new enterprises). If you can attract all newly formed companies and grow alongside them, you will become a large company as your customers grow. Companies like Stripe, Deel, Mercury, and Ramp have followed this strategy. In fact, many of Stripe's customers did not even exist when Stripe was founded.

By 2026, we will see startups that begin from scratch achieving scale across numerous enterprise software domains. They just need to build better products and fully commit to developing new customers that have not yet been constrained by existing vendors.

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。