29-05-2025
Governance Start-Ups Boom In The Battle To Keep AI Honest
Which market is growing even more quickly than artificial intelligence (AI) technology right now? It's a trick question – the answer is the market for products and services that keep AI technologies honest. The global AI governance industry was worth $890 million last year according to a recent study from Markets & Markets; by 2029, it will be worth $5.8 billion, the market research group reckons. That represents annual growth of a little over 45%.
It's not difficult to see why. 'With AI systems now making critical decisions, the potential for unintended consequences – algorithmic bias, data breaches, ethical violations – is gaining the attention of regulators and stakeholders,' Markets & Markets' analysts explain. 'Enterprises are under immense pressure to implement full governance frameworks.'
One company hoping to benefit from this driver is Unbound, a San Francisco-based start-up with a particular focus on data privacy and security. 'We operate as a gateway,' explains Rajaram Srinivasan, co-founder and CEO of the company. 'Many businesses now have AI policies, but while they ask their staff to follow those guidelines, there's not much enforcement going on; we provide a way to automate for that.'
The idea is for Unbound's customers to use it as a platform through which they plug into the AI tools they're deploying in their business. The platform then filters the way staff are using these tools; for example, it checks what data they're using, to ensure sensitive information isn't being shared with the large language models employed by AI providers. It can also be used to trial new tools and to monitor the cost of usage.
'Enterprises want visibility into what's being used, assurance that their data is protected, and the ability to swap in better models as the space evolves,' Srinivasan adds. 'Unbound is a bridge to make that possible.'
It's an idea that appears to resonate with customers. Since its launch in 2023, Unbound has signed up a number of mid-market and enterprise customers, particularly in data sensitive sectors such as health and technology.
Chief technology officer and co-founder Vignesh Subbiah, says these clients have already prevented more than 7,000 potential data leaks using Unbound tools. 'Surgical security controls in every AI request enable teams to innovate freely without putting corporate secrets at risk.'
The Hut Group, a UK-headquartered e-commerce company, is one early adopter of Unbound's platform. Abraham Ingersoll, chief information security officer of the company, says the technology enables his team to ensure staff are able to make the best use of AI tools without taking undue risk. 'We see the security team as an enabler, not a blocker,' Ingersoll says. 'Unbound empowers us to roll out AI tools to employees with confidence.'
These early customer wins have also attracted the attention of investors. Unbound is this week announcing it has raised $4 million of seed funding. The round was led by Race Capital, with participation from Wayfinder Ventures, Y Combinator, Massive Tech Ventures and a number of angel investors.
'AI is projected to reach $4.8 trillion in market value for the enterprise by 2033 globally, but without proper guardrails, that value is at risk,' says Edith Yeung, general partner at Race Capital. 'From shadow models to data leaks, the dangers of unmanaged AI are very real.'
Indeed – and a growing number of start-ups think this is a potentially lucrative area of the AI industry to target. Research published by StartUs Insights earlier this year identified more than 1,000 start-ups in the AI governance space, offering a huge variety of services. They range from risk management specialists such as FairNow in the US and Germany's Trail, to compliance solutions such as Inspeq AI in Ireland and KomplyAI of Australia.
