AI Won’t Save Journalism ... But It Might Save Journalists Who Learn to Use It

By Alastair Goldfisher
Veteran journalist and creator of The Venture Lens newsletter and The Venture Variety Show podcast. Alastair covers the intersection of AI, startups, and storytelling with over 30 years of experience reporting on venture capital and emerging technologies.

A terabyte of evidence. A first-degree murder trial. A public defender drowning in digital files.

That’s where TrialKit co-founder Ariel Deshe found himself before pivoting from practicing law to launching a tech startup. His mission at TrialKit is to build AI tools to help attorneys handle the overwhelming volume of digital discovery flooding today’s cases.

Deshe isn’t alone. Across the legal community, AI is shaking things up, upending billing models, displacing associates, raising concerns about accuracy and even triggering questions about the future of judges.

“The goal isn’t to replace attorneys. It’s to give them back time and let them focus on the work that actually needs their judgment,” Deshe told me. “AI can process a mountain of data, but the human still decides what matters.”

TrialKit, an AI-powered discovery platform designed for criminal defense attorneys, announced its public launch in early April alongside a $4.25 million seed round led by UpWest. The company provides an AI-driven legal discovery platform to help defense attorneys manage the overwhelming volume of digital evidence in modern criminal cases.

“The amount of data just exploded,” Deshe said. “And I didn’t see a way to practice law well without better tools.”

TrialKit is just one example of how AI is being put to work inside law firms. But it’s not just defense attorneys feeling the shift. On the corporate side and throughout the legal world, AI is transforming the practice of law and the business model behind it.

“There’s a huge impact,” said Louis Lehot, a Silicon Valley corporate lawyer I interviewed on The Venture Variety Show. “Law firms are a great testing ground for enterprise AI companies. We have so many processes that are great use cases, both in the practice of law and the business of law.”

Overall, AI is doing for legal eagles what it offers to professionals in multiple sectors: increase efficiency. What used to take 30 hours can now be done in 30 seconds. Generative AI is helping law firms scan data rooms, flag contract terms, analyze filings and prep due diligence, all with a few well-tuned prompts.

Here is why it matters: This isn’t just happening in law. Every industry is experiencing shifts in workflows as they fall under the speed and scale of AI. If you’re a founder, investor or platform leader, the takeaway is the same: AI isn’t just making tasks faster. It’s changing what clients value, what they’ll pay and what teams are expected to deliver.

Discovery is exploding

Deshe told me that when he was still a public defender in Colorado, it seemed like the data load exploded overnight.

“Cases that were a couple hundred pages started looking like a couple hundred gigabytes,” he said. “We were the same number of attorneys with the same tools. We couldn’t keep up.”

TrialKit uses multimodal AI tech—including transcription, vision models and language search—to pull insights from massive amounts of evidence, helping defense attorneys sort through critical facts in minutes instead of weeks​.

TrialKit is built specifically for criminal defense work, where attorneys face tight timelines, massive discovery files and limited resources. The platform handles video, audio, text and images to transcribe, categorize and make all of it searchable through natural language queries. On top of that, it includes litigation prep tools, like witness and fact charts, designed to help attorneys go to trial faster and more confidently.

Will new associates still have jobs?

AI is especially changing the work of associates. “First-, second- and third-year associates will have to find different roles. Research has become a commodity,” Alon Shwartz, co-founder of Trellis, which provides state trial court data, told me in a recent episode of The AI Cognitive Shift podcast on YouTube​

Yes, that’s a problem. And it’s not because the work of associates can’t be replaced, but because their learning curve can’t. Law firms are wondering how the next generation will gain experience if the next generation skips the foundational work AI now does for them in seconds.

Deshe echoed that concern. “It’s a real question I hear. What happens to new attorneys who never learned to go through discovery page by page?”

It’s not just productivity that’s on the line. Across law firms and legal departments, AI is forcing a rethink of how services are staffed, priced and delivered.

The end of hourly billing?

Another change occurring in the legal community, and which is likely to take place in other professions, as well, is that the standard billable hour is under pressure.

More than 70 percent of U.S. law firms now offer fixed fees, capped rates or outcome-based pricing. While hourly billing still exists, its dominance is waning as clients are refusing to pay top dollar for AI-assisted work, such as document review and legal research.

“It’s a real tension,” Shwartz said. “Clients know what AI can do. They’re asking, ‘Why is this taking so long?’”

As gen AI accelerates this transition, Thomson Reuters projects AI could save 266 million lawyer hours annually, pressuring firms to stop charging fees for time spent.

Here’s some math to consider. The American Bar Association reported there were about 1.33 million US. lawyers as of 2023. If AI reduced each lawyer’s billable hours by 4 hours per week, or let’s say about 200 hours per year, then at an average rate of $500 an hour, that’s a loss of $100,000 in billable hours per attorney each year that AI has erased. Great savings for clients. Not so good for the bottom line of legal firms.

Additionally, nearly half of legal professionals predict AI will further erode hourly billing within five years, as clients demand transparency and refuse to subsidize efficiency gains, according to Thomson Reuters’ Future of Professionals Report.

Big bets on legaltech boom

With the pressure mounting on traditional business models, it’s no surprise that investors are paying attention to the disruption as funding is flowing toward AI-first legaltech startups.

Several legaltech companies have made headlines for their AI tech, and they’ve raised large venture funding rounds to match. Harvey recently closed a $300 million Series D round, signaling just how hot the AI-for-law space has become.

Luminance, which automates document review and contract analysis, recently raised $75 million in a Series C. And Eudia, a company using AI agents to handle routine work for in-house legal teams, locked in $105 million earlier this year.

What connects all these legaltech companies is that they’re aiming to solve how to make legal work faster and cheaper and to scale without sacrificing quality or judgment.

And venture capital will flow to whichever company can help the legal industry strike the balance between speed and accuracy, while also proving that AI isn’t replacing lawyers, but making the best ones more valuable.

Meanwhile, legal firms are adapting. Some use AI-generated reports to highlight issues and loop in attorneys only where human insight adds value. That’s leading to more work, not less, according to Lehot.

“I find that AI tools actually increase demand for lawyers,” he said in the clip below. “I find that when you’re able to be more specific with clients and show them things that you otherwise wouldn’t have been able to because of the cost, they want to know more.”

Listen to more insights from Lehot in a full podcast episode on The Venture Variety Show as he talks about VC-backed exits, investor behavior and smart moves founders can make in a murky market.

The hallucination problem

Of course, there are risks with using gen AI. Deshe and Shwartz each flagged the issue of hallucinations, where AI makes up responses that sound plausible but are 100% false.

Examples of AI creating fake case law are cropping up all too regularly, including a recent personal injury case in Wyoming involving Walmart, as Reuters reported.

“In law, you don’t want creativity,” Shwartz said. “You want accuracy.”

That’s why tools like TrialKit, and others built with similar use cases in mind, are careful to link every AI-generated response directly back to the source material.

And then there’s the broader fear of whether judges are next in line to go the way of the dinosaur?

Shwartz speculated on the idea of AI-powered courtrooms. “Do you want an AI judge? Maybe it’s more impartial,” he said. “But that opens a whole new debate.”​

You can hear more perspectives from Shwartz on the impact of AI on the legal community in an episode of The AI Cognitive Shift podcast on YouTube.

Where this is headed

One appellate attorney told Deshe that AI is already doing a better job than he can, at least in spotting legal issues. “It writes better briefs, and it takes five minutes instead of five months,” the attorney told him. “It will absolutely replace my job.”

But Deshe said the attorney didn’t sound bitter. More like resigned​. Deshe calls this the beginning of a bigger shift.

“It won’t eliminate lawyers,” he said of AI. “But it will definitely change what they do.”

And for small firms or solo lawyers, AI might level the playing field. Consider this, much like AI allows a sole startup founder to do more, AI also makes it easier for smaller firms to litigate big claims, take on more clients and move faster without needing an army of staffers.

“I think AI will definitely change what lawyers do,” Deshe said. “I don’t think it will eliminate lawyers.”

Law is just one example of an industry facing the pressure of scale, speed and expectation. If AI is rewriting the legal playbook, how’s it impacting your industry, too?​

🎙️ Stay informed by subscribing to The Venture Lens for the latest insights and updates from Alastair.