Opening business doors to LLM powered agents
In the early 2000s, businesses scurried to establish an online presence or risk irrelevance (1). Then came the mobile revolution, putting…
In the early 2000s, businesses scurried to establish an online presence or risk irrelevance (1). Then came the mobile revolution, putting brands literally in consumers’ pockets (2). Now, another seismic shift is on the horizon: the rise of Large Language Models (LLMs) (3). Soon LLMs will be capable of interacting with businesses. This post explains why and how to take advantage of this new paradigm.
Why LLMs matter for business
My hypothesis is that as people grow more comfortable conversing with intelligent agents, we’ll see agent-to-business interactions becoming the standard user interface (UI) for both B2B and B2C operations. In this context, businesses will develop their products and services to be accessible via APIs, allowing intelligent agents like Luzia to easily interact with them.
Building on this, the advancements in speech-to-text technology — such as Whisper and some of META’s latest models (3) — present a massive opportunity to increase the accessibility and universality of these services. Voice interfaces could become the natural next frontier for user-agent interactions. Imagine a world where a simple spoken command to Luzia not only triggers complex agent-to-business interactions but also makes these services more universally accessible, reducing the barriers that currently exist with multiple-step online processes.
This vision isn’t just mine; technology leaders are echoing similar sentiments. As Bill Gates recently stated:
“Whoever wins the personal agent, that will be a big thing because you’ll never go to a search site again. You’ll never go to a productivity tool again. You’ll never go to Amazon again. Everything will be mediated through your agent”
Opening Doors to LLMs: What’s In It for Businesses?
Imagine a version of Luzia that not only helps you plan your weekly menu but also orders your groceries. Or one that recommends vacation spots and books your stay. This is not sci-fi; it’s the future we’re building. By the way, this is not new; the Economist spoke about it in 2020; it has just accelerated more than expected (4).
Incorporating intelligent agents like Luzia into their ecosystems opens up new avenues for businesses. One of the most immediate advantages is the likelihood of increased business volumes through frictionless transactions (think about Amazon’s one-click purchase). By streamlining the consumer journey — from product discovery to purchase — intelligent agents can dramatically reduce drop-off rates and boost conversions. Additionally, the coordination between different business actors, facilitated by agents, can significantly enhance operational efficiency and even create new, integrated service offerings.
Moreover, the inherent accuracy and reliability of intelligent agents will likely lead to lower failure rates, reducing the need for customer support and thereby cutting costs. This is part of a broader trend towards automation that not only improves efficiency but also allows businesses to reallocate resources to more strategic initiatives. So, it’s not just about making life easier for consumers; it’s a win-win scenario where businesses can operate more effectively while offering a better user experience.
Navigating the Challenges Ahead
Technical Hurdles
Agents today struggle with managing multiple integrations and basic tasks (5), leading to clunky user experiences. Moreover, feature-rich agents consume LLM resources at a high rate, which at this point are both expensive and scarce. While these challenges are formidable, they are not insurmountable and there will be both proprietary technology and industry resources poured into this problem. For instance, we’ve already tackled the issue of LLM consumption by developing a proprietary, context-aware solution in which Luzia enables tools as needed. Furthermore, as I discussed in this post, the speed of innovation is such, that what now seems impossible, may be a few lines of code in a couple of months. Seriously, I know I keep saying this, but a year ago chatGPT didn’t even exist!
The Need for Standardization
Imagine a world where each business uses a different language for its website, rendering web browsers useless. Similarly, a myriad of integrations necessitates some form of standardization at both the agent and API level.
In the evolving landscape of LLMs, we foresee a platform and a standard where businesses can easily open their APIs or expose their individual agents. Two are leading the pack at this point, OpenAI plugins and Langchain Tools.
But remember, this isn’t about one agent to rule them all; it’s about creating a robust ecosystem where various agents can interact with each other to provide their users with a multitude of services, all without having to create custom-made integrations. Each user might still have their favorite agent, but now that agent can tap into a much wider range of options, all standardized, through a single, powerful platform. Again, this requires standardization.
Imagine the potential for businesses to access hundreds of millions of consumers that would interact effortlessly with their products and services. I am optimistic.
The Future of Luzia
Consider this conversation:
BK: “Luzia, book me a table at a nice Indian restaurant in Madrid tonight at 7 pm.”
Luzia: “You have a conflicting meeting. Should I reschedule?”
BK: “Yes, move it to tomorrow.”
Luzia: “Done. Your table at the Taj Mahal Palace is booked. I’ll notify Allison that the meeting with her will now be tomorrow at 5 pm.”
This is the (NEAR) future we’re working towards at Luzia. We aim to simplify complexities, offering a layer of abstraction through language. As we sail into the era of LLMs, Luzia isn’t just riding the wave; we’re navigating the currents, reshaping how businesses function in this new age.
The future is here, and it speaks the language of AI. If you own a business and want to join our waitlist for integrations, send me a DM