What are the latest funding trends in AI governance?

All funding deals in the AI governance market over the last years

Below is the table listing all the deals. You can find our methodology at the end of this page.

If you want a deeper understanding of the market and its current dynamics, get our report covering the AI Governance Market.

Is AI governance funding actually booming right now?

AI governance funding is growing very fast right now, and the increase runs much deeper than the headline dollar figure.

From January 1 to September 21, 2026, we tracked 48 comparable financings across 46 companies. Over exactly the same period in 2025, there were 33 deals across 31 companies. Deal activity is therefore up 45.5%, while the number of funded companies is up 48.4%.

Capital has moved even faster. The companies in our dataset raised $1.134 billion, versus $397.7 million over the comparable 2025 period and $189.4 million two years earlier. The median round also moved from $7 million last year to $10 million today.

More companies are getting funded, and investors are putting more money into the typical financing at the same time. That makes the current acceleration much harder to dismiss as a temporary spike caused by a couple of oversized rounds.

AI governance funding measure	2026 YTD	2025 same period	2024 same period
Deals	48	33	22
Unique companies	46	31	22
Capital raised	$1.134B	$397.7M	$189.4M
Capital excluding >$50M rounds	$539.5M	$297.7M	$189.4M
Median round	$10.0M	$7.0M	$6.0M
Average round	$24.1M	$12.8M	$8.6M
Largest round	$125M	$100M	$27.5M
Follow-on share	58.3%	42.4%	54.5%

Is the AI governance boom just a few giant funding rounds?

No. Even after removing every AI governance round above $50 million, 2026 funding is still up 81% from the same period last year.

The large rounds are obviously helping. Zenity raised $125 million, Onyx Security $113 million, HiddenLayer $100 million, Neo $75 million, Straiker $64 million, Arcade.dev $60 million and WitnessAI $58 million.

Yet $539.5 million remains once every >$50 million financing is stripped out, compared with $297.7 million a year earlier. The median round is also up 43%, which is hard to explain with megarounds because medians barely care about the biggest deals.

Concentration has actually come down. The three largest financings account for 29.8% of current capital, versus 45.3% last year. So there are more very large rounds now, but the market depends less on its top three transactions than it did in 2025.

The $1.134 billion headline is magnified by exceptional rounds, but it is not being created by them. The underlying financing market is much stronger too.

Are more AI governance startups getting funded, or are investors simply writing bigger checks?

AI governance is getting both broader and more heavily funded: we are seeing more companies raise money and larger checks across much of the market.

The company count increased from 31 to 46, almost exactly alongside the rise from 33 to 48 deals. If repeated rounds for a tiny set of startups were driving the boom, deal count would be pulling away from company count. It is not.

Round sizes are changing at the same time. The median increased from $7 million to $10 million, and the average went from $12.8 million to $24.1 million.

The distribution makes the shift easier to see. Deals below $5 million fell from 36.4% of activity to 18.8%. Meanwhile, we found ten rounds between $20 million and $50 million, compared with only three a year ago.

The middle of the market is much thicker now. Companies no longer need to reach a $100 million megaround for the financing environment to look dramatically different from 2025.

Deal-size bracket	2026 deals	2026 share	2025 deals	2025 share
<$5M	9	18.8%	12	36.4%
$5M–<$20M	20	41.7%	14	42.4%
$20M–<$50M	10	20.8%	3	9.1%
$50M–<$100M	5	10.4%	1	3.0%
$100M+	3	6.3%	1	3.0%
Median round	$10.0M	—	$7.0M	—
Average round	$24.1M	—	$12.8M	—
Reliable deal amounts	47	97.9%	31	93.9%

Are Seed-stage AI governance startups getting squeezed out?

Seed funding is still alive in AI governance today; the change is that Series A, B and C activity is growing much faster around it.

We counted 29 Seed deals this year, up from 25 during the same part of 2025. So there is no broad retreat from early-stage companies.

What changed is the layer above Seed. Series A moved from five rounds to 11, Series B from one to four, and we now have two Series C financings where the comparable 2025 period had none.

Series A alone attracted roughly $401 million. Series B and C together now account for only 12.5% of deals but about 42% of usable capital.

That is a big maturity shift, and it is happening while the absolute number of Seed financings is still rising. AI governance is adding a scale-up market on top of its early-stage base.

Are AI governance investors mostly backing existing winners now?

Existing AI governance companies are taking a much larger share of the money now, although new startups are still entering the market at roughly the same pace.

We identified 20 first financings in the current period, compared with 19 last year. There is very little change there.

Follow-on activity doubled from 14 rounds to 28. The median follow-on financing then jumped from $11 million to $30 million, while the median first financing only moved from $6.5 million to $7.5 million.

Companies such as Straiker, WitnessAI and Geordie show what this looks like in practice. Straiker went from a $21 million financing in 2025 to a $64 million Series A in 2026. WitnessAI followed its earlier $27.5 million Series A with $58 million this year. Geordie moved from a $6.5 million Seed to a $30 million Series A within roughly nine months.

Roughly 86% of usable current capital now sits in follow-on financings. The new-company pipeline is still there, but most of the extra dollars are going into companies that investors have already backed once.

Financing measure	2026 YTD	2025 same period	2024 same period
First financings	20	19	10
Follow-on financings	28	14	12
First-financing share	41.7%	57.6%	45.5%
Follow-on share	58.3%	42.4%	54.5%
Capital in first financings	$156.8M	$148.4M	$79.8M
Capital in follow-ons	$977.7M	$249.3M	$109.7M
Median first financing	$7.5M	$6.5M	$5.5M
Median follow-on	$30.0M	$11.0M	$7.05M

Which AI governance categories are winning today?

AI Policy Enforcement and AI Governance Platforms are where the funding market has moved most aggressively.

Together they account for 38 of the 48 current deals and roughly $951 million of capital. Last year, the same two categories produced only 18 comparable financings and around $235 million.

Policy Enforcement has gone from 12 deals to 22, with capital rising from $180.6 million to $636.7 million. Governance Platforms moved from six deals to 16 and from $54.1 million to $314.1 million.

We can see the same direction in the companies raising the largest rounds. Zenity is building controls around autonomous AI activity. Arcade.dev focuses on what enterprise agents are allowed to execute. Runlayer combines agent deployment with permissions and control. Willow gives agents identities, scoped access and runtime guardrails.

Investors are putting much more weight on products that sit directly in the path of an AI system's actions, rather than products that only document what happened afterward.

AI governance category	2026 deals	2025 deals	2026 capital	2025 capital
AI Policy Enforcement	22	12	$636.7M	$180.6M
AI Governance Platforms	16	6	$314.1M	$54.1M
AI Assurance Services	4	8	$81.0M	$120.8M
AI Audit Software	2	5	$38.0M	$34.3M
Governance Evidence Tools	3	1	$60.1M	$1.4M
AI Compliance Tools	1	1	$4.5M	$6.5M
Policy Enforcement + Governance Platforms	38	18	$950.8M	$234.7M

Is AI assurance falling out of favor with investors?

Standalone AI assurance has clearly lost funding breadth this year, even though a few assurance companies are still raising large rounds.

AI Assurance Services fell from eight deals to four, taking its share of activity from 24.2% to 8.3%. AI Audit Software also dropped from five financings to two.

The category has not disappeared. Gray Swan raised a $40 million Series A, while Coval raised $28 million for testing and monitoring autonomous voice agents.

What has changed is where assurance sits inside the product. Coval, for example, covers simulation, live monitoring and continuous evaluation as enterprises move voice agents into production. Gray Swan has also expanded toward runtime detection for policy violations.

Investors still care about testing and assurance, but lately the better-funded products tend to connect those capabilities with continuous production controls.

Why is runtime AI governance getting so much money now?

Runtime AI governance is getting funded because enterprise AI increasingly has permission to do things. Controlling those actions has become an infrastructure problem, not just a model-quality problem.

Zenity's $125 million financing is one of the clearest examples. The company describes enterprises putting agents into finance, engineering, HR, customer service and security workflows, where the systems can make decisions and execute actions.

Arcade.dev made essentially the same issue central to its $60 million Series A. Its product checks whether an agent acting for a particular user is actually allowed to perform a specific action on a specific resource.

Willow provides another concrete example at a much earlier stage. Before its $7 million Seed round, the company's access layer was already running inside Wix with roughly 5,000 weekly active users across teams including engineering, product, HR, finance and legal.

Then there is Straiker. When it raised $64 million, the company reported that run-rate revenue had grown more than 15 times in less than a year as enterprise demand for agent security accelerated.

These companies approach the problem differently, but the commercial pressure behind them looks similar. Once AI can send a message, change a record, access a system, write code or move through a workflow on someone's behalf, enterprises need identity, permissions, monitoring and enforceable rules around those actions.

Is the EU AI Act actually helping drive AI governance funding?

The EU AI Act is probably helping AI governance demand, especially in Europe, but regulation alone cannot explain the current funding boom.

The timing is unusually relevant. Obligations for general-purpose AI providers started applying on August 2, 2025. The European Commission's enforcement powers for those obligations started on August 2, 2026, including the ability to impose fines.

European funding activity has strengthened at the same time. We tracked ten European deals in the current comparable period, versus only three last year.

There are now funded governance companies across the United Kingdom, France, Denmark and Italy. Geordie raised $30 million to govern enterprise agents, AI Score raised $5.4 million after reporting rapid commercial growth in regulated and data-sensitive industries, and White Circle raised $11 million around real-time monitoring and control of AI activity.

Still, the biggest current financings also revolve around agent deployment, security and production control outside Europe. Regulation looks more like an extra push on an already-growing operational problem than the single reason investors are pouring money into the sector.

Is North America losing its lead in AI governance?

North America is losing share of AI governance deal count, but it still dominates the big-money end of the market.

North America's share of deals fell from 75.8% last year to 54.2% today. Europe rose from 9.1% to 20.8%, the Middle East from 12.1% to 16.7%, and Asia-Pacific from 3.0% to 8.3%.

Capital tells a very different story. North America still takes about 80% of current funding, and its median round has jumped from $7 million to $28 million.

Europe's growth comes mainly from having many more funded companies. Its median is only around $6.3 million. Asia-Pacific has also broadened from one deal to four, while still representing little more than 1% of capital.

Startup formation is spreading out quite quickly. Large growth financings remain heavily concentrated in North America for now.

Why are corporate investors suddenly showing up in AI governance deals?

Corporate investors are becoming much more active because AI governance is moving closer to the systems their own customers and employees actually use.

Strategic investors participated in 15 current deals, up from five during the comparable period last year. Their share of financings has increased from 15.2% to 31.2%, and the median round involving a strategic investor has risen from $10 million to roughly $35 million.

The names are revealing. Morgan Stanley and Wipro invested in Arcade.dev. Citi Ventures and Workday Ventures participated in Straiker. M12 and Booz Allen Ventures joined HiddenLayer. Qualcomm Ventures and Samsung Ventures backed WitnessAI. Zenity's round included Hitachi Ventures, LG Technology Ventures and Intel Capital.

These companies have direct exposure to enterprise software, cloud infrastructure, regulated industries, workplace systems or large-scale AI deployment.

Corporate participation has moved well beyond small experimental Seed bets. It now appears regularly in Series A, B and C rounds around the control infrastructure enterprises expect to need themselves.

Is the AI governance funding market controlled by the same few VCs?

AI governance funding is drawing in a wider investor base, rather than simply getting more money from the same small group of specialists.

We found 51 unique disclosed lead or co-lead investors in the current period. The equivalent number last year was 35, and two years ago it was 26.

That increase of roughly 46% is almost identical to the growth in deal volume.

A few names recur. Bessemer Venture Partners appears three times as a lead or co-lead, while SYN Ventures, Hetz Ventures, Norwest and Ugly Duckling Ventures appear twice each.

But there is no single syndicate sitting behind most of the expansion. Lead-investor breadth has kept pace with the number of financings, so interest in AI governance is spreading through the venture market rather than staying inside a closed circle of dedicated funds.

Has AI governance already raised more in 2026 than in all of 2025?

Yes. AI governance had already passed the entire audited 2025 total in both deals and dollars before the current period ended.

Full-year 2025 produced 41 comparable deals and approximately $489.9 million of usable capital. The 2026 period covered here has already reached 48 deals and $1.134 billion.

That puts current funding at around 2.3 times the entire 2025 capital total, while deal count is about 17% higher than all of last year.

Compared with 2024, the gap is even larger: full-year capital that year was about $210.6 million.

We should not turn those comparisons into an automatic full-year forecast because venture funding arrives unevenly. We do not need to. The current year has already exceeded the previous full year without making any assumption about what happens next.

Is AI governance turning into an AI security market?

AI governance and AI security are overlapping much more heavily now, especially wherever autonomous agents can access tools, data and production systems.

HiddenLayer's $100 million Series B is explicitly tied to agentic runtime protection, including autonomous coding agents. Straiker describes itself around discovering, testing and protecting enterprise agents. WitnessAI is expanding agentic governance and security capabilities. AIR Security focuses on the skills and add-ons agents can use.

Yet the market still contains classic governance platforms, compliance tools, evidence systems and assurance products, so collapsing everything into cybersecurity would hide meaningful differences.

The cleaner read is that governance is moving closer to execution. The more control a product has over permissions, identity, runtime behavior and actual agent actions, the more naturally its work overlaps with security.

That convergence helps explain why Policy Enforcement has become the biggest category in our current dataset.

What is actually changing in AI governance funding right now?

AI governance funding is going through a real step-change: more companies are raising, normal rounds are larger, and a scale-up layer that barely existed a year ago is now taking a large share of the capital.

The broadening is easy to see. Deals increased from 33 to 48 and funded companies from 31 to 46. The typical round also became larger, with the median moving from $7 million to $10 million.

The rise survives a very aggressive outlier test. Removing every round above $50 million still leaves funding 81% higher than last year. The growth cannot be reduced to Zenity, HiddenLayer or another individual company.

The biggest structural change sits in follow-on financing. First financings barely moved, from 19 to 20, while follow-ons doubled from 14 to 28. Investors are still funding new entrants, but much more of today's money is going to companies that have already proved enough to raise again.

Category rotation is equally clear. Policy Enforcement and Governance Platforms now account for almost four-fifths of all deals, while standalone assurance and audit have lost share. Runtime control, agent authorization, permissions and continuous governance are where investors are putting the most weight.

Current company announcements also give us a plausible reason. Zenity, HiddenLayer, Arcade.dev, Willow, Straiker, Geordie and Runlayer all describe variations of the same enterprise problem: AI agents are moving from experiments into real workflows, and companies need a reliable way to decide what those agents can see and do.

The part we should treat much more cautiously is company age. A large share of the founding-year field in our raw database contains a systematic 2009 error, leaving too few reliable observations to make a serious age-based claim.

Overall, AI governance funding today looks both broader and more mature than it did a year ago. The startup base is still expanding, while the most dramatic growth is happening one or two financing steps later, as runtime-governance companies begin raising genuine scale-up rounds.