Language Model UIs Threaten to Disintermediate Apps

Public software suppliers and private asset firms such as Blue Owl or Ares Capital now face investor turbulence caused by concerns about the impact of language models on enterprise software. 

The enterprise software segment of the market has lost about 30 percent of its value over the past three months or so, for example. And private investment firms that have moved into software as a service find investors questioning the stability of dividends that typically fuel buying of such assets. 

For decades, software value was exemplified by the Graphical User Interface humans used to interact with data.

As language models  become the "front-end" of the enterprise, they threaten to hollow out traditional apps,  as the AI query box becomes the primary interface where work actually happens, with the value of the traditional apps shifting to the the backend database functions. 

When a user can type, "Find all overdue accounts in the Northeast and draft personalized follow-up emails," they bypass the customer relationship management app. 

That same substitution arguably challenges many other enterprise app functions, in at least some instances and for some use cases. 

Traditional App Category	Legacy UI Workflow	LLM/Agentic Substitution
Business Intelligence (BI)	Navigating complex filters, SQL queries, and static dashboards.	Conversational Analytics: "Show me a chart of Q3 churn by region compared to last year."
CRM / Sales Tools	Manual data entry, lead scoring menus, and email template selection.	Autonomous Sales Ops: "Draft a follow-up for the ACME lead based on our call notes yesterday."
HRIS / Onboarding	Employee portals with nested forms for benefits and documentation.	Employee Concierge: "I need to update my 401k and check my remaining PTO."
IT Service Mgmt (ITSM)	Filing tickets, selecting categories, and waiting for manual routing.	Self-Healing Desk: "My VPN is down; run the standard reset and let me know when I can reconnect."
Project Management	Moving cards on a board, manual status updates, and Gantt chart shifts.	Project Orchestrator: "Update the project timeline based on the delay in the design phase."
ERP / Finance	Reconciling line items across spreadsheets and procurement modules.	Agentic Finance: "Match these 50 invoices to their purchase orders and flag any discrepancies."

Some might argue that the application software suppliers most affected include vendors whose apps primarily add a user interface to someone else's model.

When the underlying "infrastructure" (the model) begins to support app-layer functions, the value of apps that merely "pass through" that intelligence drops close to zero.

Writing and grammar tools were among the first to be hit. Grammarly, Jasper, Copy.ai, and specialized "AI essay or email" writers are in this category.

When your operating system can proofread, rewrite, and change the tone of an email directly in the text box, why do you need a third-party tool?

So basic creative writing and editing are no longer "premium" products; they are "hygiene features" of the operating system.

The second category at risk are "aggregation" layer suppliers including traditional travel aggregators such as Expedia and Booking.com; basic SEO-driven content sites, and "Search-as-a-Service" tools.

These companies built business moats based on human-centric search friction. When an AI agent can be a substitute, the value of a user-friendly "comparison portal" evaporates.

Headshot generators, stock photo sites, and simple "background remover" apps likewise are being replaced by features integrated directly into messaging and design suites as well as language models. 

So which firms and apps are more safe? Those which have access to private sources of key data; apps that “own” customer data and apps offering key regulatory compliance features. 

Defensibility Pillar	Description	Example of a "Safe" Company
The Data Loop	Do they have data that only they can access (and that isn't on the public web)?	Glean (internal corporate knowledge graphs) or Bloomberg (proprietary financial data).
The System of Record	Is the company where the "final truth" of a business resides?	Salesforce (Agentforce). They own the customer data; an AI model is useless without being "grounded" in that specific record.
Regulatory Moats	Does the app operate in a space where "safety" and "compliance" are harder than the "intelligence"?	Veeva (Life Sciences) or specialized legal AI platforms that handle chain-of-custody and HIPAA compliance.

The threat is often not so much outright and full replacement, but a diminution of sales volume. If an AI agent can perform the work of five junior analysts using a specific tool, the enterprise requires fewer "seats" or licenses.

The “brand value” implications also exist. When the user never logs into the actual app, but only uses the query box, the app loses its "stickiness." 

Also, as AI automates code generation, enterprises are able to build custom, lightweight "wrappers" over their own data.

AI App Layer Moves Illustrate How Disruptive New Tech Erases Industry Boundaries

Enterprise software equities are said to have lost as much as $1 trillion in market value over the past six or so trading days, as there is concern that language models moving into the application layer of software will successfully substitute a natural language query process for full enterprise software supporting business intelligence, enterprise workflow and function applications (graphics, inventory management, knowledge management, for example). 

Anthropic's Claude features new plugins and features integrated into "Claude Cowork" for tasks in legal, sales, marketing, and data analysis, for example.

These allow AI to automate workflows traditionally handled by specialized SaaS platforms from companies like Salesforce, Workday, SAP, and ServiceNow. 

Many argue this represents a fundamental challenge to the software-as-a-service business model. OpenAI is making a similar push with "Frontier," an AI agent platform for orchestrating tasks across corporate systems.

This trend exemplifies how disruptive technologies often dissolve long-standing boundaries between industries, allowing newcomers from one sector to invade and redefine another. 

Disruptive Technology	Original Domain	Disrupted Industries	Impact
Printing Press (c. 1440)	Publishing/Mechanical Engineering	Scribes, Religious Institutions, Education	Ended elite control over information, spawning bookselling and accelerating the Renaissance, but bankrupting scribes.
Steam Engine (1760s–1840s)	Mechanical Power	Transportation, Manufacturing, Agriculture	Fused energy with industry, urbanizing societies and disrupting horse-based transport, but creating factory jobs.
Electricity/Combustion Engines (1870s–1910s)	Energy Generation	Manufacturing, Communication, Automotive	Blurred energy with production, disrupting horse carriages and enabling global trade, while phasing out older crafts.
Diesel Locomotives (1930s)	Engine Technology	Railroads, Steam Engineering	Erased lines between automotive tech and rail, slashing travel times (e.g., Burlington Zephyr's 1,000-mile non-stop run) and antitrust scrutiny for GM.
Personal Computers/Internet (1970s–1990s)	Computing/Electronics	Media, Retail, Communication	Dissolved barriers between tech and commerce (e.g., Amazon's rise), disrupting print media and creating digital economies.
Smartphones (2000s)	Mobile Telephony	Computing, Photography, Entertainment	Blurred consumer electronics with software, bankrupting Kodak and Nokia while birthing app ecosystems.
Cloud Computing (2000s)	IT Infrastructure	Retail, Data Storage	Erased e-commerce vs. infrastructure, disrupting hardware firms like Dell and enabling startups.
Ride-Sharing Apps (2010s)	Software/Mobile Tech	Transportation, Taxis	Fused digital with mobility, upending taxi monopolies but creating gig economy debates.
Streaming Services (2010s)	Digital Media	Entertainment, Cable TV	Blurred content creation with distribution, eroding cable bundles and empowering creators.
Electric Vehicles (2010s–2020s)	Battery Tech/Automotive	Traditional Auto, Energy	Dissolved auto with renewable energy, pressuring Ford/GM and reducing oil dependence.

Disruptive innovations typically begin by targeting underserved markets or offering simpler, cheaper alternatives, then over time scaling to challenge incumbent value propositions more directly. 

Will AI "Eat Enterprise Software?"

If you are an investor in enterprise software, you are aware there is a fear that language models are going to disrupt the traditional enterprise software market and firms. And that fear seems to be playing out in equity prices.

At one level there is concern that the traditional pricing model (per-seat licenses) will be challenged.

At another level there is concern that increasingly-capable AI models will displace the need for many enterprise software functions. 

Investors are essentially moving from views that “software eats the world” (so invest) to “software is dead” (so stay away). Near-term turbulence is inevitable. 

But it also is possible to argue that long term, there will be more enterprise software, even as AI adoption accelerates. 

More to the point, though language models enable natural language interfaces, automate routine tasks and generating insights from vast datasets, they arguably cannot replace enterprise software. 

Enterprise systems are engineered for reliability, security, scalability, and regulatory compliance in high-stakes environments. Moreover, enterprises often deal with proprietary data silos, strict data privacy laws and mission-critical uptime that general-purpose models cannot easily replicate. 

Aspect	Role of Enterprise Software	Role of General-Purpose Models	Coexistence Example
User Interface and Interaction	Provides structured dashboards, forms, and workflows for consistent, role-based access.	Enables natural language querying and conversational interfaces for ad-hoc exploration.	Models integrated as chatbots within enterprise resource planning systems (querying inventory via plain English without navigating menus).
Data Management and Security	Handles secure storage, compliance (audit trails, encryption), and integration with legacy databases.	Analyzes unstructured data or generates summaries, but relies on external data feeds.	Enterprise tools feed sanitized data to s for insights, while maintaining control over sensitive information ( GDPR-compliant AI assistants in CRM).
Automation and Workflow	Executes rule-based, repeatable processes like approvals or batch processing with high reliability.	Automates creative or variable tasks, such as generating custom reports or code.	Models suggest workflow optimizations within HCM platforms, but the core execution remains in the enterprise system (auto-drafting performance reviews in Workday).
Analytics and Insights	Delivers predefined key performance indicators, business information tools, and real-time dashboards with deterministic accuracy.	Provides probabilistic predictions, trend spotting, or scenario modeling from natural language prompts.	Hybrid BI where enterprise software runs core analytics, and models enhance with exploratory queries ("What if" simulations in financial planning tools).
Customization and Scalability	Supports enterprise-grade customization via APIs, modules, and cloud scaling for thousands of users.	Offers flexible, on-demand generation but struggles with consistent scaling or versioning.	Models used to generate custom code snippets for enterprise integrations, deployed within the platform (auto-building plugins for Salesforce).
Compliance and Auditing	Ensures regulatory adherence with built-in logging, versioning, and certification	Lacks inherent auditability; outputs can be opaque or inconsistent.	Enterprise systems log  interactions as auditable events, using AI for efficiency while meeting standards (fraud detection in banking software).

It’s a bit analogous to the traditional choices between general-purpose and application-specific processing. Sometimes one makes more sense than the other, but both coexist. 

AI-enabled or AI-centric software is moving up the stack of what a product is. So consumer experiences of products include vastly more software content than in prior years.

Sometimes a general-purpose approach will suffice. But not always. ASICs still make sense as well. 

And AI will often allow software to become more capable, rather than replacing it, which is the common concern today. 

Domain experience, codified in enterprise software, arguably will be just as important tomorrow as it is today. 

But investors, at the moment, seem more focused on the near-term negative impact on enterprise software company fortunes. 

In some cases, that concern is exacerbated by huge new capital spending requirements for AI infrastructure. 

An adage suggests "markets can stay irrational longer than you can stay solvent." And that is the reality some investors might be facing in the near term.