AI Meets M&A: Insights From a Dealmakers' Bootcamp

For years, mergers and acquisitions (M&A) has relied on relationships, rigorous due diligence and expert judgment. Now, generative AI (GenAI) is starting to add real muscle to that equation. But this isn’t about technology replacing people — it’s about helping deal teams work more efficiently so they can focus on high-value tasks.

That was the focus of a recent AI bootcamp for corporate development dealmakers led by legal thought leader Ari Kaplan and hosted at Pillsbury’s offices in Silicon Valley, where senior professionals from corporate development, legal and IT teams came together to share firsthand how GenAI is being tested and adopted in live deal environments. The message was clear: the technology is already having an impact — and the most successful teams are using it to solve targeted problems.

Where AI is developing early wins for dealmakers

Early use cases include drafting letters of intent (LOI), redlining agreements and analyzing past deal terms to accelerate repetitive tasks and improve decision-making speed. Unsurprisingly, due diligence stood out as the most active area for AI integration. Attendees described how artificial intelligence (AI) is being used to parse large volumes of contracts, compliance documents and financials — automating responses to standard questions and surfacing red flags faster.

Solutions like Intralinks’ AI-powered DealCentre AI are part of this tectonic shift in dealmaking. The platform has been purposefully designed to enhance how dealmakers interact with deal-related data by surfacing relevant insights and automating Q&A during due diligence. Instead of combing through hundreds or thousands of files manually, deal teams can now ask questions directly and receive precise, contextual answers drawn from the data room itself.

Importantly, these advances are being pursued with a cautious, structured approach. AI tools are undergoing rigorous vetting, data security remains paramount and many organizations are implementing responsible AI programs and mandatory training to ensure safe and effective use.

Post-merger integration (PMI) is another area ripe for disruption. Attendees discussed how AI could help teams map organizational structures, identify technology overlaps, monitor sentiment and track synergies — bridging the often-fragmented handoff from diligence to integration.

The biggest takeaway? GenAI is already changing how deals get done. The challenge now is to scale adoption strategically — balancing innovation with oversight and empowering teams to use AI as a true force multiplier.

Final thoughts

Our AI bootcamp underscored one critical takeaway for dealmakers: the future of M&A will be a partnership between human expertise and AI-driven efficiency. By complementing traditional practices with cutting-edge technology, corporate development teams can unlock new levels of insight and value creation in the dealmaking process.

Speaking on an episode of The Dealist podcast, Intralinks’ Head of AI Analytics Prakash Kanchinadam noted, “Humans remain central to decision-making, especially for critical or material aspects of a deal. AI’s role in those processes is to empower — not replace.”

Corporate deal teams are actively seeking practical, low-risk opportunities to test GenAI. The most effective approach? Start with a clear, repeatable challenge — like non-disclosure agreement (NDA) review, Q&A routing or contract summarization — and apply GenAI to solve that specific problem. This issue-first, solution-forward mindset is what turns curiosity into capability — and pilots into long-term performance gains.

As tools like DealCentre mature and adoption becomes more widespread, the competitive edge will go to those deal teams who implement AI with purpose and rigor. The opportunity is real — but so is the responsibility to deploy it thoughtfully.

Transcript

Welcome to the Deep Dive. Today we're really digging into something big, generative AI and how it's changing corporate dealmaking. Yeah, it's a hot topic. We've got some great insights actually from a recent AI bootcamp. Lots of M&A pros were there.

Right. And the goal here is to give you, the learner, a quick handle on the key shifts happening. We're pulling out the essential stuff from talks involving corporate development, legal, IT folks, basically people who are right in the thick of it, either using AI now or seriously looking into it for their deals.

So think of this as getting the inside scoop. We're using notes from that bootcamp, some audio clips too.

Exactly. It paints a picture of where AI is being used now in M&A, what's working, what the hurdles are.

And um, the real game-changing potential. From finding targets right through to integration.

That's the mission then — distill that practical knowledge for you.

Let's do it.

Okay. So first things first, where is this generative AI actually, you know, making inroads in the deal space right now?

Well, the bootcamp feedback was interesting. It seems like product development, customer service, engineering, they're kind of leading the charge. Corporate development, it's lagging a bit, but uh, interest is definitely picking up fast.

Okay, so corp dev is playing catch up but getting serious. Are there specific things they're using it for?

Yeah, absolutely. Attendees specifically brought up using generative AI for drafting and negotiating, like letters of intent, the actual M&A agreements, stock purchase agreements. Stuff like that.

So really getting into the core legal documents.

Exactly. It points to a pretty targeted use. Trying to find efficiencies in those really, um, time-intensive parts of a deal. Think about, you know, getting a first draft of an LOI done quickly using past deals as a base. Frees up the humans for the trickier bits.

Precisely. And it sounds like these tools aren't just being adopted blindly. Internal AI teams are playing a big part.

Oh, definitely. That came through loud and clear. These internal teams are crucial for vetting new tech. And the motivation is pretty straightforward. Boosting efficiency, finding smarter ways to do the work.

Makes sense. What's the general feeling about, say, outside lawyers? Is AI going to replace them?

Not entirely, no. The consensus seems to be more that, uh, generative AI won't make outside counsel obsolete, but it's definitely expected to, you know, bring down the spending on them.

Okay. Companies are thinking they can handle more routine tasks in-house with these AI tools, potentially shaving off, say, 10–15 percent on standard diligence tasks.

Right, which frees up budget for more strategic advice.

Exactly. Now, here's something interesting from the bootcamp. Apparently, when they did a show of hands, most people said their companies are officially using generative AI.

Yeah, but often in like very specific controlled ways.

How so?

We're talking things like internal sort of sanitized versions of ChatGPT. Or maybe protected co-pilot setups. It's cautious, you know, very focused on control and security first.

Which is totally understandable, right? Given how sensitive M&A data is.

Absolutely. And that ties into the discussions they had about maybe building their own proprietary large language models using the company's historical deal data.

That's a big undertaking though.

Yeah, I bet. One person described the challenge like finding a pedometer solution, something versatile enough for all their different historical data, but, uh, not too simplistic. Finding that sweet spot.

But it wasn't all about building custom stuff, was it? There was a specific tool mentioned. NotebookLM.

Ah. Yes, NotebookLM got a lot of attention. For anyone who hasn't heard of it, it basically lets you upload a bunch of documents. I think someone said up to 300.

Okay.

And then the AI helps you pull out insights from those documents. Someone from one firm was really enthusiastic. Said they used it for negotiating LOIs, figuring out typical escrow percentages by feeding it past M&A deals.

So it analyzes your own data.

Exactly. And that's key. Its big selling point is that it treats your uploaded data as the source of truth. That really helps cut down on the risk of AI just, you know, making things up — hallucinations.

That's crucial. What else do they like about it?

Well, apparently, the interface is pretty intuitive. And it can do cool things like generate audio summaries, almost like little podcasts about your documents, plus executive summaries, timelines, mind maps, FAQs — all based only on what you uploaded.

And you can share these notebooks.

Yeah, internally and externally. So good for collaboration too. It's using natural language processing to really understand and boil down the info you give it.

Interesting. Were other tools mentioned beyond NotebookLM?

Yeah, a few. eBay talked about using their own custom GPT called HubGBT. For searching internal and external info.

Okay.

And then in the legal tech space, you heard names like GCI, Insight, which helps with redlining agreements, and Insight for more heavy-duty legal research and analysis.

So it's a mix then, general AI tools and these really specialized legal ones.

Exactly. And that mix kind of reflects where different companies are at. The feedback suggested many started with AI in, say, product development or customer service. And now they're bringing it into internal productivity for teams like Corp Dev and Legal.

And the economic climate, is that playing a role?

It seems like it, yeah. That uncertainty is definitely pushing companies to look harder at these AI-driven efficiencies, you know, ways to tighten the belt and streamline things.

But not recklessly — you mentioned vetting. Responsible AI programs are important.

Hugely important. That was a constant theme. Companies know they need strong governance and thorough vetting for any AI tool. The potential downsides are clear if you don't have that oversight.

And training. Are people getting trained on how to use this stuff?

Yes, definitely. Some companies have pretty extensive training programs already. We heard about dedicated AI teams with ambassadors to help spread the knowledge, even mandatory AI training for some roles. It's all about making sure people can use these tools effectively and responsibly.

Or skilling up the workforce.

Mm-hmm.

Which brings us nicely to looking at how AI fits into the actual stages of a deal. Starting with, uh, deal flow and pipeline management.

Okay, the very beginning. Finding the targets.

Yeah. And interestingly, the sense from the bootcamp was that generative AI is maybe used less here currently compared to, say, due diligence or integration.

Why do you think that is?

Well, that early-stage stuff — identifying targets, managing the pipeline — it's often really driven by the corp dev team's expertise, maybe with some legal input. It's traditionally been very relationship-based, too.

True. But there's interest in using AI here too.

Oh, for sure. People are looking at how generative AI research tools could supplement the usual methods, you know, internal research, banker relationships.

How might that work?

The potential is really in analyzing huge amounts of unstructured data. Think social media buzz, industry reports online, maybe even pitch decks that come in.

Right.

The idea is maybe using AI to help rank startups, score targets on different factors, financials, IP, culture — just making that initial filtering more efficient.

But for now, it sounds like it's still mostly the traditional mix. Internal research, bankers, insights from business unit leaders.

Yeah, that seems to be the main approach still. AI in deal flow feels a bit more informal, maybe experimental at this point. Still heavily reliant on those human networks and insights.

Okay.

Though some folks mentioned using open research AI tools to explore adjacent markets, find potential targets they might not have thought of. But they also caution that, you know, accuracy can be spotty.

You still need that fundamental market understanding. You can't just blindly trust the AI output.

Not at all. Human expertise is still critical there.

Okay, so deal flow is kind of nascent, but due diligence and closing sounds like that's where the action is right now with generative AI.

Absolutely. That came across really strongly. Due diligence was flagged as a primary focus area for AI adoption by the attendees.

And diligence is usually a big cross-functional effort, right?

Yeah, often involves lots of departments, sometimes even a dedicated diligence team running point for the whole deal.

So where does AI fit in?

The big potential everyone sees is transforming that Q&A process. Imagine using the data room itself — all those documents — as a private large language model. So it could potentially answer a huge chunk of the standard due diligence questions automatically just by analyzing the data that's already there.

That could save a massive amount of time. What kind of specific tasks?

Things like quickly summarizing financials or complex IP filings, analyzing contracts to flag key risks or clauses, highlighting compliance or litigation worries, even spotting market trends or competitor red flags buried in the data.

Basically, finding the needles in the haystack much faster.

Exactly. It's about surfacing the crucial insights from just vast amounts of information.

You mentioned eBay earlier. They had a concrete example.

They did, and it was pretty impressive. They're essentially treating their diligence data room as an LLM to answer thousands of standard questions. It works. They reported really high accuracy in their tests, something like 98%, and said it saves literally hundreds of hours. Their integration team actually uses it to clean up the question lists, deduplicate, and get answers out faster, reducing the burden on the target company too.

Wow, that addresses a huge pain point, doesn't it? That slow, frustrating back and forth on Q&A when the answer's probably already in the data room somewhere.

Totally. There's a real hunger for tech that can just streamline that whole process and make sure the right info gets to the right people without all the chasing.

But people are still relying heavily on outside law firms for diligence.

Oh yeah, that's still very common. But what's changing is that companies are starting to expect more transparency from their law firms. About what tools they're using. You know, are they leveraging AI like Kira or Harvey, which are known tools in the legal space.

Right. Wanting to know if their advisors are using the efficiency tools, too.

Exactly. Exactly.

And, of course, you still have those internal debates slowing things down, sometimes concerns about data security, responsible AI use, especially with super sensitive deal data.

Understandable. Controlling the data, avoiding IP leaks — that's paramount.

Absolutely. That risk is a major consideration when you think about sharing data externally, even with advisors.

Is AI being looked at for risk assessment itself?

Yeah, that was mentioned, too. Exploring how Gen AI could help quantify M&A risks, maybe even automatically generate risk heat maps and keep those models updated as new info comes in during the deal.

So a more dynamic way to track risk.

Potentially, yes.

Yeah. Moving towards a more data-driven approach there.

Okay, let's shift gears to the phase after the deal closes. Post-acquisition integration. Sometimes overlooked, but so critical.

Absolutely critical. And yeah, it's typically another cross-functional team effort — commercial, sourcing, legal, finance people — usually with a dedicated integration lead coordinating everything.

What are the challenges there that AI might help with?

Well, one common issue is that handoff from diligence to integration. It's often, let's say, suboptimal. Knowledge gets lost. Information doesn't flow smoothly.

Right, silos.

Exactly. So, AI could potentially help bridge that. Imagine using it to quickly map out the two organizations' structures, their tech stacks, to spot overlaps or integration challenges early on. Maybe analyzing employee sentiment from internal comms to gauge morale during the transition, monitoring KPIs automatically to track synergy realization, helping ensure compliance with things like earnouts or regulatory filings post-close.

Lots of potential applications there, too.

For sure. The feeling though was that there's maybe a need for better enterprise-level AI tools for integration. Ones that go beyond just legal applications. Tools that support the whole integration effort.

Right. And again, it often comes back to solving that question problem, making sure people in the newly combined company can get answers easily, breaking down those knowledge barriers.

And ultimately tying it all back to the deal's original goals.

That's the aim — linking the synergies you achieve back to the why of the deal, and making sure that translates into how the combined company operates, its KPIs, and so on.

So pulling this all together, what are the big challenges and where are things heading?

Well, a key need that came up is for better, more integrated enterprise AI platforms with strong backing from internal IT, not just point solutions.

Makes sense. A more holistic approach.

Yeah. We also saw that adoption speed really varies. Some companies are just naturally more conservative about new tech than others. Culture plays a big role.

And it's not just about the tech itself, is it?

Not at all. A really crucial point was that successful adoption hinges on solving a real problem for people. Making them feel like the tech actually helps them do their job better. It has to feel relevant.

It's about the people and the process, not just the tool.

Precisely. Which means overcoming internal resistance, getting that buy-in, is still a major hurdle sometimes.

The advice was, start with the problem you need to solve, then see how AI fits — not the other way around.

And building empathy. Understanding why people might be hesitant.

Yeah, that seemed really important. Understanding concerns, facilitating change, bringing people along on the journey. That's key for making AI adoption stick.

But despite the challenges, there's optimism.

Definitely. There's a sense that we could see some pretty big strides in AI for dealmaking over the next, say, few quarters. Mainly because the tools themselves are getting better, more user-friendly, more targeted.

And the end goal remains the same.

Pretty much. Use AI strategically to streamline the whole deal process, get major efficiency gains, and ultimately unlock more value from M&A.

Okay, so to wrap up our deep dive, what are the main takeaways?

I think the big picture is that generative AI is making real inroads in corporate dealmaking. Due diligence seems to be the main beachhead right now. But there's huge potential being explored in pipeline management and post-acquisition integration too.

But adoption is still a bit patchy.

Yeah, it's growing, but unevenly. Companies are wrestling with important things like data security, getting the right internal support structures in place, and just figuring out the best strategic way to implement this stuff.

So for you, the learner, hopefully this gives you a solid overview of the current trends, the practical ways AI is being used, and how it's starting to reshape the M&A landscape.

Right. Which leads us to maybe a final thought to chew on. Given how fast AI is evolving, what are the real long-term implications here? How might the roles of corp dev, legal, all the functions involved in M&A actually change?

That's a big question.

It is. How will that balance between human smarts and AI automation continue to shift? Could we see entirely new roles, new kinds of expertise becoming essential in dealmaking down the road, specifically because of AI?

Definitely things to think about.

We really encourage you to consider how these insights apply to your own work and keep exploring this rapidly changing intersection of AI and M&A.

Absolutely. Thanks for joining us for this deep dive.