Wedbush: AI intelligent agents will not replace enterprise software platforms

LootboxPhobia · 2026-02-25T02:41:29+00:00

Wedbush analysts believe that concerns about AI tools replacing existing enterprise software are exaggerated. The case presented by Anthropic shows that the practicality of AI tools is limited by data access and cannot replace a complete software ecosystem. AI agents need to rely on existing workflow systems rather than replace them. Analysts point out that cybersecurity will become even more important in this environment.

LootboxPhobia

2026-02-25 02:41:29

Abstract generation in progress

Investing.com – Wedbush analysts on Tuesday said that after Anthropic hosted an enterprise AI event showcasing Claude Cowork features, concerns that AI tools will replace existing enterprise software systems have been overstated.

During the event, Anthropic demonstrated several real-world applications across different industries. Spotify reduced engineering time for complex code migrations, Novo Nordisk improved clinical research documentation timelines, and Salesforce shortened cycle times in Slack. The company noted that these AI tools are only useful within the scope of accessible data and will not replace existing software ecosystems.

The analysts stated that the market is confusing the capabilities of foundational models with full enterprise software replacement. Demonstrations by Anthropic and OpenAI show model intelligence but lack workflow orchestration, compliance infrastructure, auditability, security controls, integration, billing systems, uptime guarantees, and enterprise-level service agreements.

Companies like Microsoft (NASDAQ: MSFT), Salesforce (NYSE: CRM), ServiceNow (NYSE: NOW), and Pegasystems are deeply embedded workflow record systems. The firm said replacing them would require removing critical task infrastructure, not just adding a large language model.

The analysts noted that foundational models exist within workflow engines rather than replacing them. AI agents need an orchestration layer to operate across systems, which existing platforms already provide.

Wedbush stated that AI agents and autonomous workflows increase the attack surface through more APIs, machine identities, lateral movement risks, and cloud-native workloads. This expansion requires more runtime monitoring, identity governance, model security, and zero-trust execution.

The firm believes CrowdStrike (NASDAQ: CRWD), Palo Alto Networks (NASDAQ: PANW), and Zscaler (NASDAQ: ZS) will be winners in this environment, highlighting cybersecurity as the execution layer for AI.

The analysts pointed out that Anthropic and OpenAI lack the 20-year enterprise distribution networks, CIO relationships, and embedded vertical workflows that companies like Salesforce, ServiceNow, and Microsoft possess. They said the commercialization of model layers will outpace that of workflow layers.

Wedbush stated that the urgency of AI is accelerating deal cycles and driving modernization efforts, rather than bypassing existing installed bases. The firm added that AI is reducing friction in traditional transformation projects.

This article was translated with the assistance of artificial intelligence. For more information, please see our Terms of Use.

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.