The beginning of the end of the transformer era? Neuro-symbolic AI startup AUI announces new funding at $750M valuation
The buzzy but still low-key startup in New York City Augmented Intelligence Inc (AUI)that aims to go beyond the popular ‘transformer’ architecture used by most contemporary LLMs such as ChatGPT and Gemini, has raised $20 million in a SAFE bridge round with a valuation cap of $750 million, bringing its total funding to nearly $60 millionVentureBeat can exclusively reveal.
The round, which closes in less than a week, comes amid heightened interest in deterministic conversational AI and precedes a larger raise that is now in late stages.
AUI is based on a fusion of transformer technology and a newer technology called ‘neuro-symbolic AI’, described in more detail below.
“We realize that you can combine the brilliance of LLMs in linguistic capabilities with the guarantees of symbolic AI,” said Ohad Elhelo, Co-founder and CEO of AUI in a recent interview with VentureBeat. Elhelo launched the company in 2017 alongside co-founder and Chief Product Officer Ori Cohen.
The new financing includes participation from eGateway Ventures, New Era Capital Partners, existing shareholders and other strategic investors. It follows a $10 million increase in September 2024 at a $350 million valuation cap, which coincided with the the company’s announced go-to-market partnership with Google in October 2024. Early investors include Vertex Pharmaceuticals founder Joshua Boger, UKG chairman Aron Ain and former IBM president Jim Whitehurst.
According to the company, the bridge round is a precursor to a significantly larger increase that is already in an advanced stage.
AUI is the company behind Apollo-1, a new foundational model built for task-oriented dialogue, which it describes as the “economic half” of conversational AI – distinct from the open-ended dialogue handled by LLMs like ChatGPT and Gemini.
The company argues that existing LLMs lack the determinism, policy enforcement and operational certainty that businesses need, especially in regulated industries.
Chris Varelas, co-founder of Redwood Capital and advisor to AUI, said in a press release to VentureBeat: “I’ve seen some of today’s top AI leaders walk away with their eyes rolling after interacting with Apollo-1.”
A distinctive neuro-symbolic architecture
Apollo-1’s key innovation is its neuro-symbolic architecture, which separates language skills from task reasoning. Instead of using the most common technology underlying most LLMs and conversational AI systems today – the vaunted transformer architecture described in the seminal 2017 Google paper ‘Attention Is All You Need’ – AUI’s system integrates two layers:
-
Neural modules, powered by LLMs, handle perception: encoding user input and generating natural language responses.
-
A symbolic reasoning engine, developed over several years, interprets structured task elements such as intentions, entities and parameters. This symbolic state machine determines the appropriate next actions using deterministic logic.
This hybrid architecture allows Apollo-1 to maintain state continuity, enforce organizational policies, and reliably trigger tool or API calls – capabilities that transformer-only agents lack.
Elhelo said this design emerged from a multi-year data collection effort: “We built a consumer service and recorded millions of human-agent interactions across 60,000 live agents. From that, we abstracted a symbolic language that defines the structure of task-based dialogues, regardless of their domain-specific content.”
However, companies that have already built systems around transformer LLMs need not worry. AUI wants to make adoption of its new technology just as easy.
“Apollo-1 is deployed like any modern foundation model,” Elhelo told VentureBeat in a text message last night. “It does not require dedicated or proprietary clusters. It works in standard cloud and hybrid environments, uses both GPUs and CPUs, and is significantly more cost-efficient to run than frontier reasoning models. Apollo-1 can also be deployed across all major clouds in a separate environment for greater security.”
Generalization and domain flexibility
Apollo-1 is described as a basic model for task-oriented dialogue, meaning it is domain agnostic and generalizable across industries such as healthcare, travel, insurance and retail.
Unlike consulting-heavy AI platforms that require building custom logic on a per-customer basis, Apollo-1 allows enterprises to define behaviors and tools within a shared symbolic language. This approach supports faster onboarding and reduces maintenance in the long term. According to the team, an enterprise can launch a working agent within a day.
Crucially, procedural rules are encoded at the symbolic layer and not learned from examples. This enables deterministic execution for sensitive or regulated tasks.
For example, a system might block the cancellation of a Basic Economy flight not by guessing intent, but by applying hard-coded logic to a symbolic representation of the booking class.
As Elhelo explained to VentureBeat, LLMs are “not a good mechanism if you are looking for certainty. It is better if you know what you are going to send [to an AI model] and always send it, and you always know what will come back [to the user] and how you approach that.”
Availability and access for developers
Apollo-1 is already in active use within Fortune 500 companies in closed beta, and a wider general availability release is expected before the end of 2025, according to a previous report of The information, which announced the first news about the startup.
Companies can integrate with Apollo-1 through:
-
A developer playground, where business users and technical teams jointly configure policies, rules and behavior; or
-
A standard API, using OpenAI compatible formats.
The model supports policy enforcement, rules-based customization and control via guardrails. Symbolic rules allow companies to dictate set behavior, while LLM modules handle open text interpretation and user interaction.
Enterprise Fit: when reliability wins over fluidity
While LLMs have advanced general-purpose dialogue and creativity, they remain probabilistic – a barrier to enterprise deployment in financial, healthcare and customer service industries.
Apollo-1 addresses this gap by providing a system where policy compliance and deterministic task completion are first-class design goals.
Elhelo says it clearly: “If your use is a task-oriented dialogue, you should use us, even if you are ChatGPT.”
