AI Gets Down to Business: DISCO, Amwell and Doximity Give Wall Street Something to Measure -( $AMWL $DOCS $LAW )

Legal software growth, digital-health discipline and a new medication-safety tool point toward a more practical technology opportunity.

Artificial intelligence has become exceptionally good at commanding attention. The more interesting question is which companies can turn that attention into useful products, durable customer relationships and better financial results. CS Disco (NYSE: LAW), American Well Corporation (NYSE: AMWL) and Doximity (NYSE: DOCS) offer three distinct approaches. DISCO is pairing legal software growth with an early move into positive adjusted EBITDA. Amwell is tightening its operating model and raising its annual outlook. Doximity is adding a medication-dosing verification layer to its clinical AI platform, addressing a problem that cannot be solved with a more confident-sounding answer. Together, they illustrate a potentially rewarding shift from AI spectacle toward execution: technology embedded in professional work, designed to solve identifiable problems and supported by measurable business milestones. The companies are not at the same stage. Doximity already operates a profitable business, while DISCO and Amwell are pursuing more consistent profitability. That distinction makes the group more interesting, not less. These are different routes toward the same commercial objective: becoming useful enough that customers keep coming back.

DISCO’s Q3 2026 Results Give Legal AI a Financial Foundation

DISCO reported third-quarter 2026 revenue of $45.3 million, up 11% from a year earlier. Software revenue reached $38.6 million, increasing 10%, and management cited its sixth consecutive quarter of double-digit software growth. The underlying comparison was stronger after accounting for an unusual prior-year contribution. Excluding revenue associated with a contingent legal matter in the third quarter of 2025, total revenue increased 15% and software revenue rose 13%. Those figures provide context for the reported growth rates rather than replace them. The more consequential milestone arrived further down the income statement. DISCO generated approximately $0.1 million in adjusted EBITDA, reaching that threshold one quarter earlier than management had anticipated. Its GAAP net loss narrowed to $5.0 million from $13.7 million. A six-figure adjusted EBITDA result will not require additional vault space. It does, however, mark a useful transition from discussing when a business might cross break-even to assessing whether it can stay there. The comparison requires some accounting discipline. The prior-year GAAP loss included substantial stockholder-litigation expenses. On an adjusted EBITDA basis, the improvement was more measured, moving from a loss of roughly $0.3 million to a slightly positive result. Both developments matter, but they tell different stories.

Larger DISCO Customers Strengthen the Platform Argument

DISCO ended September with 374 customers generating more than $100,000 in revenue over the preceding 12 months, up 15% from a year earlier. Its 2025 annual report recorded 330 such customers at year-end. The comparison periods differ, but the figures show an expanding base of substantial customer relationships. That matters because DISCO is selling more than a single-purpose application. Its SEC filing describes a platform spanning legal hold, legal requests, electronic discovery, document review and case management. Multiple products create opportunities to deepen a relationship across successive stages of legal work. The commercial logic is attractive: help a legal team preserve evidence, investigate it, organize the facts and prepare its case, and the software has more opportunities to become part of the daily routine. There is still room to improve. DISCO’s dollar-based net retention was 98% at the end of 2025, a historical measure that did not yet establish sustained expansion across its existing customer base. The larger-customer count and continued software growth are encouraging, but stronger customer economics remain a milestone to demonstrate.

DISCO’s Advanced Research Gives AI a Specific Assignment

DISCO’s AI strategy becomes more persuasive when described through the work it is designed to perform. In August, the company announced the general availability of Advanced Research, an agentic AI application for multistep investigations of litigation evidence. According to DISCO, the system can evaluate initial findings, identify gaps and conduct further searches. Intended outputs include investigative memoranda, deposition outlines and timelines. Its product and training materials emphasize investigation across large datasets, visibility into the research process and citations back to documents in the customer’s database. These are company-described capabilities, not independent guarantees of accuracy. The potential advantage is the connection between the answer and the evidence. Legal teams need more than fluent prose. They need a defensible record. An AI-generated paragraph without supporting documents may be an impressive demonstration. It is a less impressive companion at a deposition. For DISCO, the opportunity is to make AI an extension of the evidence and workflows already inside its platform. The company also reported a second cohort entering the pilot program for its unified litigation solution, adding another potential avenue for broader customer adoption.

Doximity’s DoseCheck Targets a High-Stakes AI Weakness

Doximity brings the same emphasis on verification into a different professional setting. On October 29, the company announced DoseCheck, a free safety layer within Doximity Ask that checks AI-generated medication-dosing claims against a structured drug database. The system adds a rule-based verification step using FDA labeling and professional guidelines, rather than relying solely on a language model’s generated response. That distinction is central to the product’s appeal. A language model generates a likely answer. DoseCheck is designed to check whether the dosing claim conforms to an authoritative reference. In the company’s testing of nine widely used AI models across more than 8,000 dosing questions based on physician queries, at least 2.3% of dosing claims were incorrect. Doximity classified 43.8% of the detected errors as serious or potentially lethal. The company reported that DoseCheck reduced dosing errors by more than 99% in its evaluation. Those results are company-reported findings, not proof that real-world errors have been eliminated. The identified problems included incorrect units, misplaced decimal points, confusion between similarly named drugs and failures to account for patient characteristics such as kidney function, age or weight. In medicine, a decimal point is not a stylistic preference. The bullish significance is therefore not simply that Doximity has introduced another AI feature. It is that the company is addressing a specific obstacle to professional adoption: confidence that an answer has been checked before it enters a clinical workflow.

Doximity’s Distribution Gives DoseCheck a Running Start

A useful product still needs distribution. Doximity already has a substantial professional network through which to deliver it. According to its announcement, Doximity’s membership includes more than 85% of U.S. physicians. Its Clinical AI Suite, including Ask, has been deployed across more than 165 health systems, including 10 of the nation’s top 20 hospitals. DoseCheck has been rolled out to Ask users on mobile and desktop. That reach gives the new feature an existing audience rather than requiring Doximity to build one from scratch. It also strengthens the strategic case for offering DoseCheck free. The announcement does not establish a separate revenue stream for the product. Its potential business contribution is indirect: making Ask more useful and trustworthy, which could support engagement and reinforce Doximity’s position within physicians’ working lives. That is an opportunity, not a reported financial outcome. But it is a commercially coherent one. Doximity also approaches the opportunity from an established financial base. For its fiscal year ended March 31, 2026, revenue reached $644.9 million, up 13%. For the quarter ended June 30, 2026, its fiscal 2027 first quarter, revenue was $156.6 million, up 7%. Its quarterly SEC filing describes a business operating at scale with profitability. The distinction from DISCO and Amwell is important: Doximity is not waiting for AI to rescue an unproven business model. It is adding AI capabilities to an existing platform with meaningful revenue and established profitability.

Amwell’s Q3 2026 Results Put Operating Discipline in Focus

Amwell’s opportunity is different again. Its latest results support a financial-improvement story, even though a sustained revenue recovery has yet to emerge. The company reported third-quarter revenue of $49.0 million, above its previously provided guidance range. Subscription revenue totaled $23.9 million, Amwell Medical Group visit revenue was $23.7 million, and platform visits reached approximately 800,000. Reported gross margin was 53%. Management raised its full-year 2026 revenue outlook to $204 million to $206 million from $200 million to $205 million. It also improved its adjusted EBITDA outlook to a loss of $7 million to $8 million, compared with its previous expectation of a loss of $7 million to $9 million. For the fourth quarter, Amwell expects revenue of $48 million to $50 million and adjusted EBITDA between break-even and positive $1 million. Management also reiterated its objective of positive operating cash flow for that quarter. These are forecasts, not completed achievements. For many following AMWL stock, the appeal is the proximity of a measurable test. The company has placed its next financial milestone close enough that progress can be evaluated through results rather than presentation slides.

Amwell’s SEC Filings Keep the Turnaround Story Grounded

The filings also show why the bullish case should remain precise. Amwell reported revenue of $54.9 million in the first quarter of 2026, $52.0 million in the second and $49.0 million in the third. That sequence does not establish a top-line recovery. Third-quarter adjusted EBITDA also weakened sequentially, moving from a loss of approximately $1.2 million in the second quarter to a loss of $3.6 million. The latest quarter exceeded management’s expectations, but it was not an uninterrupted improvement across every measure. Cash flow provides a more constructive element. Amwell used approximately $1.0 million of operating cash in the first quarter, while reporting $9.7 million of operating cash generation for the first six months combined. The first-half statement included favorable movements in accrued liabilities and deferred revenue, while capitalized software development appeared in investing activities. Operating cash generation therefore should not automatically be treated as recurring free cash flow. The practical conclusion is encouraging without being extravagant: Amwell has demonstrated an ability to improve its cash profile. Its next task is to make that progress repeatable while stabilizing revenue.

Amwell’s VA Opportunity Offers Another Potential Catalyst

Amwell’s enterprise ambitions gained a notable development in September, when the company announced a Department of Veterans Affairs letter of intent concerning deployment of its virtual health platform within the VA enterprise. According to the announcement, the evaluation considered scalable video consultations, interoperability, cybersecurity compliance and support for integrated care delivery. Those requirements align with Amwell’s broader positioning as digital-care infrastructure rather than simply a venue for individual virtual visits. The qualification matters: the letter of intent is nonbinding. A formal agreement depends on negotiating and executing a definitive contract and meeting applicable federal procurement requirements. It is potential business, not booked revenue. Government procurement rarely shares the market’s enthusiasm for a quick timetable. Nevertheless, the opportunity fits the strategy. Amwell’s third-quarter commentary emphasized connecting clinical programs through a common platform and data structure, including a potential role in AI-enabled care. An eventual enterprise agreement could help validate that positioning, although its financial contribution remains uncertain.

Why LAW, AMWL and DOCS Belong in the Same Conversation

The connection among these companies is not that they sell the same product. They do not. It is that each is attempting to become more deeply embedded in professional work:

• DISCO connects litigation evidence, legal workflows and AI-assisted investigation. Its near-term financial test is sustaining growth while remaining at or above adjusted EBITDA break-even.
• Amwell connects digital-care programs and delivery infrastructure. Its near-term test is meeting fourth-quarter financial objectives while establishing a steadier revenue foundation.
• Doximity connects a large physician network with clinical information and AI tools. DoseCheck adds a verification function whose strategic value depends on usefulness, adoption and trust rather than an announced standalone price.

For many evaluating these stocks, that creates a differentiated set of catalysts rather than three versions of the same AI narrative. DISCO’s fourth-quarter guidance calls for revenue of $45.8 million to $47.8 million and adjusted EBITDA of zero to $1.5 million. Amwell is targeting a similar adjusted EBITDA threshold while working through a more difficult revenue transition. Doximity, meanwhile, offers an established profitable platform adding a product intended to make clinical AI safer and more dependable. Adjusted EBITDA should not be confused with GAAP net income or free cash flow. Both DISCO and Amwell still reported third-quarter GAAP losses. Their improving operating targets are meaningful, but they describe progress toward more durable economics, not the completion of that process.

The Bullish Case Is Becoming More Practical

The strongest connection across DISCO, Amwell and Doximity is the move from broad technology promises toward specific professional problems. DISCO is seeking to make litigation research faster and more useful while improving its financial footing. Amwell is working to turn digital-health infrastructure into a more disciplined business. Doximity is adding a verification layer where the cost of an incorrect answer can be far more consequential than an awkward search result. That makes the opportunity more tangible. The next developments can be judged through software growth, customer expansion, revenue stabilization, cash performance and product adoption. For many looking beyond the loudest AI headlines, these three companies offer something worth examining: technology with a defined assignment, an existing professional audience and a financial scorecard that is becoming harder to ignore.

The Sources

1. DISCO: Third Quarter 2026 Financial Results
2. Amwell: Third Quarter 2026 Financial Results
3. Doximity Company Announcement: DoseCheck Launch
4. SEC.gov: CS Disco 2025 Form 10-K
5. SEC.gov: American Well Form 10-Q for the Quarter Ended March 31, 2026
6. SEC.gov: American Well Form 10-Q for the Quarter Ended June 30, 2026
7. SEC.gov: Doximity Fiscal 2026 Fourth-Quarter and Full-Year Results
8. SEC.gov: Doximity Form 10-Q for the Quarter Ended June 30, 2026
9. DISCO: Advanced Research Launch Announcement
10. DISCO: Cecilia Q&A Advanced Research Product Overview
11. DISCO: Advanced Research Across the Litigation Lifecycle
12. Amwell: Department of Veterans Affairs Letter of Intent Announcement
13. Amwell: Care Delivery Platform and Corporate Overview