Salesforce Enhances Agentforce with Voice Capabilities for AI-Powered Customer Service

Salesforce (CRM) is doubling down on its AI-powered customer service offerings, unveiling Agentforce Voice. This new addition allows businesses to integrate lifelike voice capabilities into their AI agents, moving beyond traditional text-based interactions to engage customers in spoken conversations.

Announced ahead of this week’s Dreamforce conference in San Francisco, Agentforce Voice empowers companies to customize the AI agent’s voice – adjusting tone, speed, and even the pronunciation of specific industry terms. This level of personalization aims to create a more natural and engaging experience for customers interacting with AI-powered support systems. Crucially, the feature also allows callers to interrupt the AI agent at any point, a seemingly small detail that significantly improves user experience and prevents frustrating, one-sided conversations.

The integration of voice capabilities represents the next evolution in the generative AI boom. While the initial wave was dominated by text-based models like ChatGPT, the past year has seen significant advancements in voice AI, with companies like OpenAI and Anthropic developing chatbots capable of conducting remarkably human-sounding conversations. This technology is now poised to transform business software, promising to make customer service interactions more efficient and user-friendly.

To ensure seamless integration with existing infrastructure, Salesforce is partnering with a range of leading providers in the communications space. Agentforce Voice will be compatible with corporate phone systems from Amazon (AMZN), Five9 (FIVN), Genesys, Nice, and Ericsson’s (ERIC) Vonage.

The move comes as Salesforce faces increased competition in the AI-powered customer service market. Bret Taylor, former Salesforce co-CEO, is also vying for market share with his startup, Sierra, which launched in 2023. Sierra also offers AI agents with voice capabilities and boasts a client list including ADT, SiriusXM, and SoFi. Sierra’s rapid ascent, achieving a valuation of $10 billion, underscores the significant demand for these types of services.

Salesforce has been under pressure this year, partly due to investor concerns about disruption from AI, which could potentially reduce the need for traditional software. The stock’s underperformance relative to the Nasdaq reflects these anxieties. The broader market is increasingly aware that AI could automate a significant portion of software development tasks, leading to a potential contraction in the software engineering workforce. This looming change has spooked some investors, leading them to reassess their positions in traditional software companies.

Furthermore, the potential for large language models (LLMs) to automate tasks traditionally performed by human employees continues to be a source of concern. Recent demonstrations, like Anthropic’s Claude Sonnet 4.5 model building a Slack-like chat app in just 30 hours, have highlighted the potential for rapid displacement. Even Salesforce, in its latest earnings report, acknowledged that new AI products “may disrupt workforce needs and negatively impact demand for our offerings.” This admission, while transparent, has further fueled investor uncertainty.

Despite these concerns, Salesforce CEO Marc Benioff remains optimistic about the company’s future. He emphasizes the incremental and evolutionary nature of business, arguing that AI will augment, rather than entirely replace, existing processes. While acknowledging the potential for disruption, Benioff frames AI as an opportunity to enhance efficiency and drive innovation, rather than a zero-sum threat to the company’s core business.

Salesforce launched Agentforce last year, leveraging generative AI to handle customer requests via text-based chats. The company claims more than 12,000 implementations. However, some analysts remain skeptical about the market acceptance of Agentforce. RBC Capital Markets analysts, for example, have expressed concerns that adoption has lagged behind initial expectations for Agentforce.

Looking ahead, Salesforce plans to launch its Agent Script software in November, providing organizations with the tools to customize the AI agent’s scripts and refine its overall behavior. This move towards greater flexibility and control could address some of the concerns raised about Agentforce’s adoption rates and allow businesses to tailor the AI agent to their specific needs and brand voice.