Scared of an AI bubble? Here’s how to correctly value AI companies
As investment momentum builds, there are key challenges that investors and analysts must navigate to accurately value AI companies.
The commercialisation of AI has kick-started a global tidal wave of investment. According to Pitchbook data, 2024 saw over $315 billion ($US200 billion) of private capital funding deployed across more than 5,000 AI deals. In Australia alone, AI companies raised $1.6 billion ($US1 billion) across 48 transactions.
Despite this surge in investment, many AI firms are yet to deliver consistent revenue, let alone profit. As we know from last decade’s tech boom, it pays to get in early if you pick the right company – but how do you know if you are backing long-term value, or simply fuelling hype?
As investment momentum builds, the need for robust, evidence-based valuation frameworks has never been more pressing. There are two key challenges that investors and analysts must navigate to accurately value AI companies.
Challenge one: What’s under the hood and is it any good?
At the heart of every AI company is a complex web of technology intellectual property (IP) assets. Understanding what these IP assets are and how they work is essential to any meaningful valuation.
One key IP asset is data, with many AI models trained on vast datasets that may be proprietary, public or licensed. Its use may sit in a legal grey area, as jurisdictions consider reforms to balance innovation with copyright protection. Assessing data quality and accessibility is central to valuing it as an asset.
AI models are another core asset class. This includes algorithmic models that range from widely available open-source versions to custom-built, highly optimised proprietary models. Each requires unique considerations during the valuation process, including its relative performance and ability to be reproduced by competitors.
Similarly, specialised hardware must also be examined. The category includes custom-built chips and sensor technology designed specifically for AI model training and inference. Additional thought must be given to supply chain resilience amid current challenges in sourcing AI hardware.
Finally, AI-generated outputs which vary across industries represent valuable IP. These may include novel medical drug targets, engineered materials, or optimised industrial designs, all of which can carry significant commercial and strategic value.
Once a clear view of these IP assets is obtained, a series of reference points can be used to analyse their profile and determine their relative quality.
These reference points can be sorted under “earnings potential” or “risk profile” categories and serve as a benchmarking tool to look beyond the buzzwords, helping separate genuine breakthrough innovation from average tools dressed up to look like cutting-edge AI.
“Earnings potential” considers the predicted profitability of the underlying technology based on factors like its functional utility (or performance relative to alternatives), financial impact, market opportunity, intellectual property protection and its likely economic life.
“Risk profile” considers the technology’s development risk (including time to market, technical probability of success and technology readiness level), as well as barriers to commercialisation like political or industry challenges, and the availability of capital and resources needed to commercialise the technology.
It should be remembered that many AI companies sacrifice profitability to grow market share, raising questions about the sustainability of their business models.
Challenge two: Get the valuation framework right
Once an AI company’s assets are properly understood, it is important to apply a realistic valuation framework. Amid excitement in the AI space, some company valuations are driven by narratives and bold promises rather than verifiable results.
Further analysis of Pitchbook data shows that in 2014 the average AI or machine-learning related deal was priced at around 10 times the company’s revenue. This surged to 21 times during the post-COVID-19 tech boom. Even in 2024, it remained elevated at 15 times revenue - five times higher than the average valuation across the broader technology, media and communications (TMT) sector.
This fluctuation in deal price can be a warning sign. We’ve seen this pattern before, during the dotcom bubble, when inflated expectations led to a period of massive investment in internet-based companies between the late 1990s and the early 2000s.
This resulted in a boom in tech startup stock prices, followed by a huge crash, with many companies going bankrupt or losing most of their value. The lesson? Amid the noise, it is important to stay disciplined and realistic when assessing the underlying factors that drive enterprise value.
Specialist rating and valuation tools such as the Deloitte Tech Rating platform can be used to systematically evaluate asset value and enterprise value, allowing for better-informed investment choices.
Investors, analysts and dealmakers need a careful, structured approach when assessing companies built on new technology. By focusing on core IP assets within a clear valuation framework, they can avoid the overhype and pitfalls that have defined past tech bubbles.
Rob Burnside is Director, IP Advisory at Deloitte Australia and Christophe Bergeron is Partner and National leader, Valuations & Modelling, at Deloitte Australia.
 
Disclaimer     
 
This publication contains general information only and Deloitte is not, by means of this publication, rendering accounting, business, financial, investment, legal, tax, or other professional advice or services. This publication is not a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a qualified professional adviser.  
 
Deloitte shall not be responsible for any loss sustained by any person who relies on this publication.  
 
About Deloitte  
 
Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited, a UK private company limited by guarantee (“DTTL”), its network of member firms, and their related entities. DTTL and each of its member firms are legally separate and independent entities. DTTL (also referred to as “Deloitte Global”) does not provide services to clients. In the United States, Deloitte refers to one or more of the US member firms of DTTL, their related entities that operate using the “Deloitte” name in the United States and their respective affiliates. Certain services may not be available to attest clients under the rules and regulations of public accounting. Please see www.deloitte.com/about to learn more about our global network of member firms.  
 
Copyright © 2025 Deloitte Development LLC. All rights reserved.  
 
