Big Ideas in

American Dynamism

Bio + Health

Infra + Enterprise


Consumer Tech

Growth-Stage Tech



American Dynamism

Smart energy grids. Voice-first companion apps. Programmable medicines. AI tools for kids.
We asked over 40 partners across a16z to preview one big idea they believe will drive innovation in 2024.

American Dynamism

21st Century Public Safety

As technological development has heightened living standards, so, too, has it raised our expectations of what it means to live in a safe society. Yet many of our cities grapple with understaffed police, fire, and public safety services. While we’ve seen tech advancements benefit consumers and businesses, public safety lags far behind.

Why can’t we connect with 911 operators on FaceTime or WhatsApp to easily share photos and videos from the scene? Why do we wait for an emergency vehicle to arrive at an accident site to understand what’s happening when drones can get there faster and provide “eye in the sky” perspectives for first responders en route? According to the FBI, a car is stolen every 23 seconds in the U.S., and 40% are never recovered. Why let a crime remain unsolved when cameras and sensors can track vehicles constantly and over hundreds of miles?

Such innovations have already led to significant decreases in crime rates for early-adopter cities, but to maximize their impact, we need wider adoption. We have the means to significantly cut crime, respond to emergencies, and save lives. It’s time for a national upgrade to our public safety systems.

Smart Grids Will Power an Increasingly Electrified World

The U.S. urgently needs to revitalize energy-intensive sectors. By bypassing traditional wiring infrastructure, tech offers a solution to growing grid complexity. Distributed energy resources such as residential solar, home energy storage, and even small modular nuclear reactors not only provide reliable personal power or increased grid resiliency, but also the capability to sell surplus back to the grid. To implement this on a large scale, however, the grid must also transition from a unidirectional power flow model — from large power plants to consumers — to a “smart grid” that supports bi-directional flows from diverse sources and locations.

20% Approximate amount of electricity that comes from nuclear fission in the U.S.

Source: The U.S. Energy Information Administration

This shift in energy sources demands our most capable builders. Thus far, the transition from fossil fuels has been hampered by grid stability and the friction of integrating new power assets. Though cheap solar, wind, and natural gas have reduced electricity generation costs over the past decade, delivering that power to consumers when they need it has become significantly more expensive. The electric grid is too important to get wrong — the pervasive threat of rolling blackouts is unacceptable. At stake: Modern amenities and our industrial competitiveness.

Cost-Effective Swarming for Defense

To prepare for future conflicts, the U.S. has invested in unmanned systems such as aerial, maritime, and ground drones designed to overwhelm enemy defenses. But without cost-effective swarming, taking full advantage of overwhelming adversaries via attritable mass — unmanned aerial vehicles and other units we are comfortable losing, from an economic perspective — is limited. Swarming is the essential ingredient needed to transition the DoD’s force from one asset with many operators (an aircraft carrier, for example) to one operator controlling many assets in tandem: a true system of systems operating model. A network of autonomous systems collaborating, communicating, and coordinating would unlock a new paradigm in defense.

Software Eats Atoms via Tech-First Buyouts

Many American Dynamism companies operate in markets where value is ultimately realized in the material world. However, such companies often build their business around a core software advantage. I believe we’ll see more companies extend the scale of their software advantages by making acquisitions and implementing their software advantage post-acquisition.

There are a couple key reasons companies pursue this strategy:
1) To buy scale, in the form of operational capacity and distribution (e.g., Metropolis’ take-private acquisition of SP Plus)
2) To expand the product platform (e.g., Anduril’s acquisitions of various hardware systems)

Depending on the reason behind the acquisition, these transactions can take different forms, including the buyout of an incumbent/prospective customer, roll-ups of a fragmented market, or product-focused strategic acquisitions. For these technology-led buyouts, the common thread is that the acquirer improves the profile of the acquired business, primarily with a technology advantage. This opportunity holds particular relevance to American Dynamism companies, which often require a synthesis of software, physical capital, and operations.

Why now? There are a number of converging factors, but a particularly salient one is the power of the current AI wave to dramatically increase margins and scalability of services for operations-heavy businesses.

New Applications for Computer Vision and Video Intelligence

In 2024, we’ll likely see new applications of computer vision and video intelligence in the physical world. Leveraging insights from video data has become commonplace in the enterprise to help companies make better-informed business decisions. This capability could be even more powerful in the real world, given the richer, more comprehensive nature of the available data. However, many industries today still lack modern systems to capture and make sense of video. Customers often have no existing video infrastructure or use legacy video systems that are difficult to integrate with modern software. 

Businesses are tackling this problem by leveraging a hardware + software model, selling both the video hardware cameras as well as the software to customers. These businesses often tailor their go-to-market approach to target a very specific customer and best serve their particular needs.

Companies like Flock Safety and Ambient, for example, are both leveraging computer vision in the physical world. The same success could be found in other industries, including in transportation (e.g., roads and ports), industrials (e.g., plants and factories), agriculture, or mining. 

A New Age of Maritime Exploration

Generations ago, our ancestors took to the seas to explore. Yet today we know more about the surface of Mars than we do about our own planet’s seabed. A new age of maritime exploration is changing that, and founders are leading the way.

Maritime faces similar reliability and engineering challenges as aerospace, and many of the technologies pioneered by the latest space age can be readily applied; meanwhile, the size and importance of the commercial and defense markets provide substantial reward.

Companies like Flexport, Saildrone, Saronic, and others have already started modernizing maritime, and we anticipate that continued geopolitical, supply chain, and climate disruptions will further accelerate demand for change.

Advances in AI, hardware, and computer vision present opportunities to transform our cities, ports, and trade networks with autonomous, modernized ferries, container ships, and fishing fleets. Robots will help sustainably mine precious materials from the seafloor, map and survey waterways, and monitor the health of our ecosystems. New generations of naval and coast guard vessels, ships, and submarines will protect our supply chains and shores. Technology is once again returning to maritime. 

Bio + Health

Democratizing “Miracle Drugs”

In 2023, a wave of therapies hailed as miracle drugs — including GLP-1’s and curative cell and gene therapies — had a profound impact in patients’ lives. But our current insurance system is not set up to bear the cost of these therapies (or to accurately gauge their value, given that some are curative), nor are our healthcare providers prepared to manage the complex logistics, data collection, and clinical operations needed to realize the full benefits of these therapies. 

We look forward to seeing builders innovating at the intersection of policy, biopharmaceutical manufacturing, financing, and clinical operations so we have viable means to bring these “miracle drugs” to market without bankrupting or breaking the system.

Programming Medicine’s Final Frontier

Where are the reusable rockets for biotech? Traditional drug development is painstakingly time-consuming, risky, and expensive. It’s highly bespoke, too; one molecule has no bearing on the next molecule that gets developed — like traditional rockets, they’re one-time use only. 

That’s changing. SpaceX’s rocket reusability has transformed space travel, lowering costs and expanding horizons. Similarly, potentially curative programmable medicines like gene therapy can reuse components, like the delivery vehicles used to target specific cells, while swapping out the genetic cargo. The next mission uses the same rocket to deliver a different payload to a new destination.

The FDA is looking to the skies and taking a page out of the FAA’s approach to aviation safety — rigorous, yet adaptive — recently launching its new Office of Therapeutics Products and piloting Operation Warp Speed for rare disease to create more transparent and flexible processes for evaluating and approving programmable medicines. Imagine a future where we re-deploy, not re-invent, innovation. It will revolutionize how we make medicines, and where these medicines can take us.

Enabling Our Healthcare Workforce to Do More

As a physician-scientist, I’m always energized by new technologies that enable our existing clinical workforce to practice their superpowers at the next level. This could mean adding novel medicines to our therapeutic arsenal or developing new care delivery capabilities that fundamentally change how care teams spend their most valuable resource: time.

>50% of physicians report feeling burned-out. The most cited contributing factor was having too many bureaucratic tasks.

Source: Medscape

To this end, in 2024, I’m excited by software and data platforms that fundamentally empower providers. From ambient note-taking within EHRs, to intelligent automation of always-on triage, to precision treatment planning, there are a multitude of ways AI can reduce burnout and administrative burden and optimize physicians’ ability to provide the best and most compassionate care to patients.

Taking this idea one step further, I see a future where AI-enabled platforms could be the key that unlocks greater value-based care adoption. To date, value-based care has lagged short of its full potential, but the provider-enabling possibilities of AI could alter that trajectory for the better.

AI Will Supercharge the Future of Health

Science and healthcare have long been behind the curve on software adoption. But what was once a liability is now an opportunity, as AI leapfrogs existing software and revolutionizes healthcare tech.

Other industries that are running more efficiently on non-AI software might take more time to adopt AI. But healthcare, which is reliant on pagers, fax machines, and a bevy of humans doing manual data input, is primed for AI integration. Making this revolution easier is the fact that healthcare is the only industry with existing regulations for AI, in FDA’s regulatory framework.

In 2024, I’m looking to see these leapfrog moments taking place across the science and healthcare industry, improving the lives of providers and patients by orders of magnitude.

Consumer Tech

Voice-First Apps Will Become Integral to Our Lives

AI will finally unlock voice-first apps in the coming year, particularly in the companion and productivity categories.  

Despite being the oldest and most common form of human communication, voice has never really worked as an interface for engaging with technology — earlier this year, Microsoft CEO Satya Nadella called the past decade’s generation of voice assistants (including his company’s own, Cortana) “dumb as a rock.” Historically, people have been most likely to use smart speakers for straightforward tasks like playing music or checking the weather, not derive meaningful value from voice interactivity. Now, however, large language models have enabled virtual assistants to achieve human-level conversational capabilities. 

Importantly, voice is such a different modality for interaction that existing apps aren’t naturally equipped to build these experiences. It’s inevitable that obvious AI email features will be incorporated into Gmail, for example, but it’s less likely that Gmail will introduce an AI voice interface in your inbox. In 2024, I expect voice applications to become more useful and integrated into our lives.

Narrowly Tailored, Purpose-Built AI

In 2024, I predict we’ll see narrower AI solutions. While ChatGPT may be a great general AI assistant, it’s unlikely to “win” for every task. I expect we’ll see an AI platform purpose-built for researchers, a writing generation tool targeted for journalists, and a rendering platform specifically for designers, to give just a few examples.

Over the longer term, I think the products people use on an everyday basis will be tailored to their use cases — whether this is a proprietary underlying model or a special workflow built around it. These companies will have the chance to “own” the data and workflow for a new era of technology; they’ll do this by nailing one category, then expanding. For the initial product, the narrower the better.

AI Tools Will Teach Kids

In 2023, an estimated 30% of college students used tools like ChatGPT for schoolwork (given the nature of survey reporting, the actual figure may be higher). In the coming year, however, generative AI will begin to transform the landscape of early education.

Generative AI offers immense potential for young minds, fostering innovation and stimulating imagination. Unlike higher education, where concerns about academic integrity prevail, early education can harness AI to create a sandbox of boundless exploration. The key here lies in designing products that not only engage, but also protect our young learners. This necessitates a unique blend of content moderation, user-centric limitations, and age-appropriate interfaces. In 2024, we’re likely to see groundbreaking new AI tools, thoughtfully and meticulously designed for children. These platforms will empower children to safely take advantage of the expansive capabilities of AI and the internet.

Code-Free AI Generators Spark New Behaviors

As state-of-the-art generative AI technologies drive the marginal cost of creation to near-zero, we’ll see entirely new consumer behaviors emerge. Already, platforms like Midjourney and Ideogram allow us to create incredible images that would have previously taken hours and cost thousands of dollars. Eleven Labs can translate content in seconds — across dozens of languages — with voice cloning and audio dubbing (something I couldn’t have imagined growing up watching poorly dubbed foreign films!). Even non-developers can now stitch together a sequence of generative AI tools to create incredible outputs, without needing coding skills. Glif, for example, is a multimedia platform that lets users generate art, comics, selfies, and more with a simple prompt.

Creativity Gets Supercharged

AI creative tools narrow the gap from idea to execution. For the first time, you don’t need specialized skills and years of training to create a beautiful painting, poem, or song. However, the early products here have largely focused on the simple act of generation: creating an image, crafting an essay, or composing a track. There is immense potential for more interactive tools that act as a creative copilot and enable a true conversation with AI, far beyond the basic inpainting/outpainting functionality we see today.

For example, these products might be able to generate outputs that are editable and engage in the iterative process to refine your work. They may allow you to train models on a specific style, subject, or character to generate consistent outputs over time. Or they could help you transform existing content into something new, whether it’s animating a photo, turning a real-world video into anime, or converting a 2D image into a 3D mesh.

We’re excited about companies building the consumer UI to enable these types of experiences,  whether they’re training their own models or leveraging open-source ones.


Entering a New Era of Decentralization

As we keep seeing over and over again, when control of a powerful system or platform is in the hands of a few (let alone a single leader), it’s too easy to encroach on user freedoms. That’s why decentralization matters: It’s the tool that allows us to democratize systems by enabling credibly neutral, composable internet infrastructure; promoting competition and ecosystem diversity; and allowing users more choice, as well as more ownership. 

But decentralization has been hard to achieve in practice — at scale — when pitted against the efficiency and stability of centralized systems. Meanwhile, most web3 governance models have involved DAOs (decentralized autonomous organizations) that use simplified yet burdensome models for governance based on direct democracy or corporate governance — which are not designed for the sociopolitical realities of decentralized governance. However, thanks to the “living laboratory” of web3 over the past few years, more best practices for decentralization have been emerging. These include models for decentralization that can accommodate applications with richer features; and also include methods such as DAOs embracing Machiavellian principles to design more effective decentralized governance that holds leadership accountable. As such models evolve, we should soon see unprecedented levels of decentralized coordination, operational functionality, and innovations.

Resetting the UX of the Future

While it has been much-lamented, the fundamentals of user experience in crypto haven’t actually changed much since 2016. It’s still too complicated: self-custodying secret keys; connecting wallets with decentralized applications (dApps); sending signed transactions into increasingly many network endpoints; more. It’s more than we can expect users to learn in their first few minutes in a crypto app.

But now, developers are actively testing and deploying new tools that could reset frontend UX (user experience) for crypto in the year ahead. One such tool includes passkeys that simplify signing into apps and websites across a user’s devices; unlike passwords, which are more vulnerable and require manual work from users, passkeys are automatically, cryptographically generated. Other innovations include smart accounts, which make accounts themselves programmable and therefore simpler to manage; embedded wallets, which are built into an application and can therefore make onboarding frictionless; MPC (multi-party computation), which makes it easier for third parties to support signing without custodying users’ keys; advanced RPC (remote procedure call) endpoints that can recognize what users want and fill in the gaps; much more. All of these not only help web3 go more mainstream, but can make the UX better and more secure than in web2.

The Rise of the Modular Tech Stack

In the world of networks, one force invariably dominates all others: network effects. Network effects tend to be so powerful that there really are only two kinds of modularity — modularity that extends and strengthens network effects; and modularity that fragments and weakens them. In all but the rarest of cases, only the former ever makes sense, especially when it comes to open source. 

Monolithic architectures have the advantage of allowing deep integration and optimization across what would otherwise be modular boundaries, leading to greater performance… at least at first. But the biggest advantage of an open-source, modular tech stack is that it unlocks permissionless innovation; allows participants to specialize; and incentivizes more competition. We need more of that in this world.

AI + Blockchains Come Together

Decentralized blockchains are a counterbalancing force to centralized AI. AI models (like in ChatGPT) can currently only be trained and operated by a handful of tech giants, since the required compute and training data are prohibitive for smaller players. But with crypto, it becomes possible to create multi-sided, global, permissionless markets where anyone can contributeand be compensatedfor contributing compute or a new dataset to the network for someone who needs it. Tapping into this long tail of resources will allow these markets to drive down the costs of AI, making it more accessible. 

But as AI revolutionizes the way we produce informationchanging society, culture, politics, the economy — it also creates a world of abundant AI-generated content, including deep fakes. Crypto technology can be used here as well to open the black box; track the origin of things we see online; and much more. We also need to figure out ways to decentralize generative AI and govern it democratically, so that no one actor ends up with the power to decide for all others; web3 is the laboratory for figuring out how. Decentralized, open-source crypto networks will democratize (vs. concentrate) AI innovation, ultimately making it safer for consumers. 

Play to Earn Becomes Play and Earn

In “play to earn” (P2E) games, players would often make real-world (not just virtual) money based on their time and effort playing games. This trend is related to the broader shifts that have been transforming gaming and beyond — from the rise of the creator economy to the changing relationship between people and platforms. Web3 allows us to counter the current norm where all the proceeds of playing and transacting in games go only to gaming companies. Users spend so much time in, and generate so much value for, those platforms that they deserve to be compensated, too. 

But games were not necessarily designed to be a workplace (at least, not for the majority of players). What we really need are games that are both fun to play and that also allow players to capture more of the value they produce. As such, P2E is increasingly morphing into “play and earn”, setting an important distinction between gaming and workplace. The dynamics of how resulting gaming economies are managed will continue to shift as we see P2E evolve beyond its initial growing pains. Ultimately, however, this will not be a separate trend and will just become part of games.

When AI Becomes the Gamemaker, Crypto Offers Guarantees

As someone who spends a lot of time thinking about web3 games and the future of gaming, it’s clear to me that AI agents in games must come with guarantees: that they’re based on certain models, and that those models are executed without corruption. Otherwise, games will lose integrity. 

When lore, terrain, narrative, and logic are all procedurally generated — in other words, when AI becomes the game maker — we’re going to want to know that the game maker was credibly neutral. We’re going to want to know that that world was built with guarantees. The most important thing that crypto offers is such guarantees — including the ability to understand, diagnose, and penalize when something goes wrong with AI. In this sense, “AI alignment” is really an incentive design problem, in the same way dealing with any human agent is an incentive design problem… and is what crypto is all about.

Formal Verification Becomes Less, Well, Formal

While formal methods are popular for verifying hardware systems, they are less common in software development. For most developers outside of such hard or safety-critical systems, these methods are too complex, and could add significant costs and delays. However, smart contract developers have different demands: The systems they develop handle billions of dollars; bugs would have devastating consequences, and cannot typically be hotfixed. So there’s a need for more accessible formal verification methods in software and especially smart contract development. 

This past year, we’ve seen a new wave of tools (including ours) emerge which have much better developer experience than traditional formal systems. These tools leverage the fact that smart contracts are architecturally simpler than regular software — with atomic and deterministic execution; no concurrency or exceptions; small memory footprint and little looping. The performance of these tools is also rapidly improving by leveraging recent breakthroughs in SMT solver performance (SMT solvers use complex algorithms to identify or confirm the absence of bugs in software and hardware logic). With increased adoption of formal methods-inspired tooling among developers and security specialists, we can expect the next wave of smart contract protocols to be more robust and less prone to costly hacks.

NFTs Become Ubiquitous Brand Assets

​​More and more established brands have been introducing digital assets to mainstream consumers in the form of NFTs. Starbucks, for example, has introduced a gamified loyalty program in which participants collect digital assets as they explore the company’s coffee offerings (not to mention an AR pumpkin spice maze!). Nike and Reddit, meanwhile, have developed digital collectible NFTs that they’ve explicitly marketed to a broad audience. But brands can do so much more: They can use NFTs to represent and reinforce customer identity and community affiliations; bridge physical goods and their digital representations; and even co-create new products and experiences alongside their most dedicated enthusiasts. Last year, we saw a growing trend toward inexpensive NFTs for large-scale collection as consumer goods — often managed through custodial wallets and/or “Layer 2” blockchains with correspondingly low transaction costs. Heading into 2024, many of the conditions are in place for NFTs to become ubiquitous as digital brand assets — as Steve Kaczynski and I explain in a forthcoming book — for a wide array of companies and communities.

SNARKs Go Mainstream

Technologists have historically had the following strategies for verifying computational workloads: 1) re-executing the compute on a trusted machine; 2) executing compute on a machine specialized to the task, aka (TEE trusted execution environment); or 3) executing compute on credibly neutral infrastructure, like a blockchain. Each of these strategies has had limits in terms of cost or network scalability, but now, SNARKs (Succinct Non-interactive ARguments of Knowledge) are becoming more usable. SNARKs allow the computation of a “cryptographic receipt” of some compute workload by an untrusted “prover” impossible to forge: In the past computing such a receipt cost 10^9 work overhead over the original compute; recent advances are bringing this number closer to 10^6. 

SNARKs therefore become viable in situations where the initial compute provider can bear a 10^6 overhead and the clients cannot re-execute or store initial data. The use cases that result are many: Edge devices in the Internet of Things can verify upgrades. Media editing software can embed content authenticity and transformation data; while remixed memes could pay homage to initial sources. LLM inferences could include authenticity information. We could have self-verifying IRS forms, unforgeable bank audits, and many more uses that benefit consumers ahead.


The Rise of the Developer as Buyer in Financial Services

In 2024, the developer will become one of the most important influencers in the purchase of financial services infrastructure.

Historically, financial services infrastructure purchases were mostly driven by the economic buyer (“What’s my ROI?”) or the business lead (“Does this solve my use cases?”). But there is now a third, increasingly influential constituent: the developer. Look no further than Moov’s growing fintech developer conference and associated 4,000+ strong community. In tandem, developers at larger financial institutions and insurance companies are becoming more influential in buying decisions.  

The rise of the developer as buyer in financial services companies of all sizes favors new entrants; for fintech companies that pride themselves on a great developer experience, this will play to their advantage. Fintechs are already prioritizing the creation of developer sandboxes to let customers “try before you buy,” and are even open sourcing parts of their solutions. The developer-buyer would, of course, prefer to get a sense of how the product works prior to full implementation. But isn’t this strategy also better for everyone?

For larger financial institutions selling their infrastructure, appealing to the developer will be a new muscle that may require improvements in product architecture (including up-to-date documentation!). Still, it’s clear these institutions are recognizing the need to influence the developer, as well.

Tech Helps Community and Regional Banks Compete

1/3 of small business bank lending is provided by regional banks.

Source: BPI

In the wake of SVB and First Republic, community and regional banks are facing significant regulatory pressure and margin compression from a higher rate environment. Despite challenges, these institutions remain vital for the health and vibrancy of our local economies and small businesses. My hope is that we’ll see fintech companies step up and provide needed tools and technology to help this ecosystem of banks compete with larger institutions, manage balance sheet risks effectively, and better serve their clients. The time is now!

Financial Professional Services Get Supercharged by Software

The professional services work of financial services — the accountant, the tax adviser, the wealth manager, the investment banker — will change. These professions typically involve researching and applying learned expertise, as well as client management. Historically, software has mainly assisted in tracking workflows, with only some analytical tools (e.g. transaction categorization in accounting). With the advancement of generative AI and LLMs, more of the work can be automated, including administrative tasks, the research process (collecting and ingesting data, searching for information), summarizing and surfacing insights, and report generation. A tax advisor can more easily look through precedent to answer a question, an accountant can automatically generate a financial statement, a wealth advisor can scenario plan across a broader set of data.

Software may eventually automate the task entirely, but for now the skill set for humans in the loop will shift toward specialized expertise, reviewing the generated work, and client-facing engagement. As always, it’ll be a race between distribution and innovation: Incumbents that have already established relationships with financial professionals will need to incorporate AI into their software, while startups with modern software capabilities will need to find and build trust among new customers.

LLMs Capture New “Foundational Customer Units”

Operating systems are incredibly valuable — they own what we call the foundational customer unit (FCU). Historically, certain types of unstructured data have been challenging for operating systems to ingest. For example, Vertafore or Applied Systems in insurance have struggled to expand outside of tracking policies once they are written, since collecting data earlier in the process requires ingesting information that lives in email, PDFs, or spreadsheets. In 2024, startups leveraging LLMs will capture data that has been challenging for incumbent operating systems to collect, while automatically tagging and storing it. If these startups capture FCUs upstream from traditional platforms, we may see segments that have been served by software oligopolies shift into a new era.

New Tools for Banking and Trading 

In 2024, we’ll see ambitious founders tackle some of the messiest problems that financial institutions have to offer. 

Though the investment banking and trading services market produces nearly $350 billion of annual revenues globally, it still largely relies on systems and software built on-prem in the 1980s. While banks have started to buy cloud-based solutions to serve various horizontal needs — Salesforce for CRM, Azure for cloud computing, Databricks for lakehouse architecture — the tools vertically deployed in the banking and trading businesses to model risk, confirm/settle/clear trades, and record client orders are often manual (Excel), outdated, or both! At the same time, the buying behavior at these institutions is changing. Willingness to try new tools is at an all time high.

AI Will Push Latin American SMBs to Go Digital

Many merchants in Latin America use WhatsApp for customer service and support. These interactions predominantly involve tasks in which consumers expect quick responses, such as quoting, scheduling, and logistics. Currently, merchant response times can vary greatly depending on representatives’ availability and the complexity of the request. AI assistants could significantly streamline these time-consuming tasks, unlocking value for merchants and consumers. 

One recent example: this year, Nissan developed chatbots on WhatsApp that direct customers in Brazil to one of its nearby car dealerships. Nissan estimates that 30 to 40% of its sales leads in Brazil are now generated via WhatsApp, and the company’s average response time dropped from 30 minutes to a few seconds. 

However, though such early inroads are promising, structural challenges remain before the technology can be widely adopted by Latin American SMBs. In Brazil, for instance, over 40% of SMBs still rely on pen-and-paper operations. As automated customer service becomes increasingly prevalent and consumers grow accustomed to these prompt interactions, businesses will be compelled to digitize their workflows. There’s an opportunity for startups to facilitate this digital shift in the region.

AI Will Be the Key to Higher ROE

In the coming year, we’ll begin to see financial institutions adopt AI-native applications across a variety of operational workflows (and we’ll likely be surprised at how big of an economic impact the technology has on their P&L). While the opportunity set spans both revenue generating and middle- to back-office functions, adoption in 2024 will be focused on use cases throughout engineering, procurement, legal, compliance, and risk management.


Gaming: Where Alpha Geek Technologies Get Their Initial Product Market Fit

There are many new technologies on the cusp — AI, VR/AR, web3 — and embracing video games will be a key aspect of their success. These new technologies will change gaming, but will also be changed by gaming. For generative AI, the next stage after text and images will be 3D and video. Combined with audio, interactivity, and other facets, eventually games will cost 1/1,000th what they traditionally did to develop and will enable consumers to create their own gaming experiences. 

VR/AR has found the most product/market fit by targeting kids and teens with multiplayer gaming experiences. The next generation of headsets would do well to double down, acquiring millions of consumers in the process, rather than trying to jump to productivity tools where demand is low. For web3, each wave has been propelled by a major use case — NFTs, DeFi, etc. The next major consumer wave will come from fun, mainstream games that introduce web3 as a way for gamers to engage with the virtual items they buy.

AI-First Games That Never End

In 2024, we’ll see the first cohort of AI-first games from creators leveraging large models to enable novel game systems and mechanics. While much of the early discourse on generative AI in games has focused on how AI can make game creators more efficient, I believe the largest opportunity long-term is in leveraging AI to reinvent the nature of the games entirely — creating never-ending games that engage and retain users for a long time.

Generative agents powered by large language models will create incredibly lifelike companions and emergent social behavior, turbo-charging games with non-player characters (NPCs). Personalized character-builders and narrative systems will enable every player to have unique, personalized playthroughs of their favorite games. Game worlds themselves will no longer be rendered, but generated at run-time using neural networks. And new player onboarding will be reinvented. Every game will be designed around an AI copilot with the mantra “good alone, great with AI, and best with friends.”

Games Become “Everything Simulators”

Games are simulations that satiate our base biological primitives: for collection (Pokemon), for predator/prey (tag), for nurturing (AdoptMe), for exploration (Minecraft). Game engines facilitate the laws of the simulation. What game engines could not efficiently simulate, however, were the complex and emergent nature of human thoughts, actions, words, and goals. That is, until recently. 

Breakthroughs in LLMs and agentic frameworks now make it possible to have realistic characters in games with believable goals, actions, and dialogue. This gives the game designer a new tool in their toolbelt: the ability to simulate social dynamics. In 2024, expect to see games where the moment-to-moment involves coercion, deception, flirting, allying, leadership, peer pressure, influence, morality… you name it. Simulation tools never before possible will now be generally available.

Every biological primitive will be fair game — your innate human desire to socialize, cooperate, to find love lies at the whims of the simulation.

From Chatbots to Avatars: AI Companions Go 3D

If 2023 was the year of AI companions we text with, next year those relationships will come to life through 3D avatars we talk to in real-time, verbally. AI companion apps like Character AI are already seeing millions of monthly active users engage with chatbots like virtual Elon Musk, Super Mario, or a psychologist. In the coming year, these conversations will feel as natural as a FaceTime conversation. With lower latency responses, text-to-speech advances, and audio-driven facial animation, our conversations with AI companions will feel increasingly perceptive, present, and personalized. Entertainment will continue to shift from a passive to active experience, and the lines will blur between linear TV and interactive games.

The Next Disney is a Game Company

There’s been a lot of commentary recently on Disney getting back into video games, but I believe the next Disney will be a video game company. 

$188B Expected games revenue in 2023. By comparison, the global movie box office is expected to make $34.5B.

Source: Gower Street Analytics

2023 was a banner year for games in film and TV. The Super Mario Bros. Movie was topped only by Barbie, and The Last of Us series was the second best performing HBO series of the last decade. Though Hollywood has typically punched above its weight culturally, the global games market has never been stronger. Games are expected to gross $188 billion worldwide this year, while the global box office is only projected to reach $34.5 billion.  

As younger games-native generations play Roblox, Fortnite, Clash of Clans and Valorant, they are increasingly turning to games as their IP of choice. The reason? Games provide the deepest stories and worlds, they’re interactive instead of passive, and they offer engaging social experiences. Studios are embracing AI, which is turbocharging game creation.

When Riot Games launched the series Arcane, based on League of Legends, it became Netflix’s #1 watched and top rated series. Companies like Riot, Epic, Supercell, and new next generation game companies are poised to become the next entertainment juggernauts with games replacing films as the core of the “next Disney.” This is already happening without the mainstream noticing and will accelerate in 2024.

Anime Games Go Mainstream 

Anime has become one of the highest grossing genres, when calculated by average revenue per user. In 2022, Mihoyo grossed over $3.8 billion from releases like Genshin Impact and Honkai: Star Rail. Earlier this year, Nintendo launched Zelda: Tears of the Kingdom. In 2024, I expect anime’s momentum to continue. 

Anime is a uniquely accessible art style for kids and adults, and the medium allows for many different story archetypes. There are adventure components, romance hooks, and social loops within anime games that lead to deep player engagement. Genshin set the new standard for a fully cross-platform, performant game across multiple device types, GPUs, and frameworks. Many passionate developers are now building new experiences for players everywhere.

A New Generation of UGC Game Developers Emerges

Increasing production budgets and heightened player expectations have made it difficult to succeed as a game developer. But new user-generated content (UGC) platforms and AI-powered creator tools show promise in breaking down those barriers. 

In Q1 2023, Roblox developers earned $182 million, an estimated 17% increase from 2022. Epic has also started financially supporting Fortnite Creative developers, with projected payouts surpassing $100 million in 2023. As competition among UGC platforms grows, developers stand to benefit from greater incentives. Notably, Meta’s Horizon Worlds expanded to mobile in 2023.

Alongside better financial backing, UGC game developers now also have access to more powerful tooling driven by generative AI. (Epic has been publicly supportive of such technology and Roblox has already announced some gen AI tools.) Combined, these two factors are likely to unleash millions of new creators next year.

The Minecraft Generation Ignites the Survival-Crafting Genre

Minecraft will celebrate its 13th anniversary next year, marking a milestone as a generation of players nurtured on “crafting” games transitions into the adult gaming category. But there’s no clear game or experience that seems to support their interests. On one end of the spectrum is Rust, a fiercely competitive “crafting-survival” game-as-a-service (GaaS) behemoth first published in 2013. A more friendly offering is Valheim, launched in 2021, which smashed commercial forecasts and spurred a wave of developers to emulate its success. Many are now aspiring to create a GaaS rendition of this genre blend. The Valheim zeal just might catalyze the creation of a new, billion-dollar IP that resonates with the maturing preferences of the “Minecraft Generation.”

Growth-Stage Tech

New Modes of Storytelling, Beyond Text-Based Chat

Character.AI CEO Noam Shazeer has called entertainment “AGI’s first use case” — the ability to use AI to tell stories across text, audio, and visual formats has continued to improve at a rapid rate. In the coming year, AI will advance beyond text-based chat to multimodal models. Layering in personalization and fine-tuning by users will deepen the way we interact with AI and enable even more exciting, entertaining, and engaging experiences. It’s up to startups to create these new storytelling formats. 

CRM Will Be Powered by AI Data Collection

Sales rep data is the atomic unit of the go-to-market organization, and bad data is a problem for almost every GTM leader. No matter how many tools or plugins you throw at your CRM platform, you still have the same fundamental problem: your reps need to input accurate data. If reps input crummy data, you get crummy results.

Though we’ve seen some sales tech companies experiment with generative AI around the edges, the next generation of sales tech will use generative AI to tackle this core data problem head-on. Instead of relying on reps’ recollections or interpretations of customer meetings, these AI-native companies will be built on ground-source data, automatically captured or generated by AI, from actual customer interactions like meeting notes, emails, and call recordings.

We’ll likely see massive bottom-up adoption of these sales tools because they’ll drive significant productivity improvements. Eventually, these new AI-native sales tech companies could pave the way for a fully AI-native CRM.

The Consumer AI Battleground Moves from Model to UX

Overheard in 2023: “A minute not spent building the model is a minute wasted. Build the best model, and users will come.” 

The most popular consumer AI companies thus far have been producers of their own models, such as ChatGPT, Character, Bard, and Midjourney. Differentiation has come from being the best model in their domain: Midjourney with images, Character on entertainment, ChatGPT on overall text. UX has largely been determined by the fastest way to get models into the hands of users. 

But thanks to a combination of factors — the likelihood that chip shortages ease, the availability of most foundation models via API, and increasingly powerful open source models — the groundwork is there for breakout consumer apps to be built on someone else’s model.

In 2024, consumer AI apps will break out by delivering the best user experience around unique use cases, less on model performance alone. I’m particularly excited for consumer AI apps that figure out how to include shared experiences and a multiplayer mode, aggregate multiple models into a single interface, or build more focused solutions where workflow and process drive value. LLMs can be a source of differentiation. Today, they might provide a first mover advantage, but old-fashioned moats like network effects, high switching costs, scale, and brand are still what will likely win long-term.

Infra + Enterprise

AI Interpretability

Interpretability, which is just a complex way of saying “reverse engineering” AI models, will be big in 2024. Over the last few years, AI has been dominated by scaling — a quest to see what was possible if you threw a ton of compute and data at training these models. Now, as these models begin to be deployed in real world situations, the big question is: why? Why do these models say the things they do? Why do some prompts produce better results than others? And, perhaps most importantly, how do we control them?

Creativity Reimagined

Creativity is the most intrinsic mode of human expression, but ideas are abstract — expressing them clearly takes time and skill. Generative AI has created a path to truly democratize the means to create. From writing to painting to filmmaking, what used to take a team months to perfect now takes minutes, if not seconds. It gives everyone, skilled or not, the ability to create. 

The creative workflow has been fundamentally revamped. Prototyping and ideation are incredibly interactive. Writer’s block can now be solved by iterating with a copilot, and artistic skill sets can be honed through iteration, rather than repetition. In this new paradigm, a new set of tools are emerging that allow us to express our creativity in a multidimensional way. The key element here will be learning to compose in different modalities, including text, visual form, and audio. In 2024, these new AI playgrounds will make creative expression even more widely accessible.

B2B AI Products Embedded in Your Workflow 

Generative AI and other technologies have the potential to automate 70% of the time employees spend working today.

Source: McKinsey

In 2024, I’m optimistic we’ll see AI-native products become more embedded into workflows, performing tasks like proactively leaving comments, updating records, and completing action items after a simple approval from the user. We’re already seeing workflow-native AI products take more direct action on users’ behalf. Instead of waiting for the user to query a long document for relevant information, for example, an AI tool can proactively flag key sections.

In tandem, within B2B products I expect to see a shift away from the chat UX. Chat has been conducive in demonstrating the usefulness of LLMs, but the prompt interface ultimately disconnects the user’s workflow. In 2024, I believe we’ll see innovative AI products that are designed to meet users where they already are. 

The New “Division of Labor”

As technology evolves, it will enhance our creative and problem-solving abilities.

The big question: What will the new means of creation look like?

LLMs Advance Robotic Process Automation Systems

In 2024, I’ll be excited to see the market for LLM-powered robotic process automation (RPA) companies take off. Today, enterprises often have manual processes done on legacy software systems that are too difficult to rip and replace or to build deep integrations into. In these situations, RPA — where small “bots” are deployed to automate repetitive tasks, such as data entry — is currently the best solution. However, RPA is often still very manual and frequently breaks; it often requires a lot of custom implementation and services to stand up. 

With LLMs, there is an opportunity to build a more intelligent RPA system that can contextually understand the inputs and actions it’s taking and will be able to dynamically adjust to create a more robust solution. It’s likely there will be multiple verticalized solutions tailored to specific types of automation tasks — whether for the finance org, processing invoices, or for support org, responding to customer service inquiries — as buyers purchase the solution most tailored to their workflows and needs. 

go to top