Meta Introduces AI Agent for Business Operations
LONDON – Meta Platforms has recently launched a new artificial intelligence agent designed to assist businesses with their everyday operations. This move is seen as a strategic entry into the enterprise AI market for the well-known social media company.
The announcement was made at the company’s Conversations conference in London, where they emphasized that this new tool builds upon existing business messaging services. The agent boasts capabilities to perform tasks on behalf of businesses, such as scheduling appointments and finalizing sales.
According to Meta, over 1 million businesses have already utilized previous chatbot versions on platforms like WhatsApp and Messenger. The latest iteration will also be integrated into Instagram and will be available to businesses worldwide, regardless of size.
This initiative suggests Meta’s intent to rival companies like OpenAI, Anthropic, and Google’s Alphabet in the realm of enterprise applications, leveraging its extensive social media platforms to encourage companies to streamline their advertising and operational processes.
Naomi Gleit, Meta’s head of product, shared her insights in an interview, confirming, “This is definitely an enterprise play.” Following this announcement, Meta’s stock experienced a more than 3% increase in morning trading.
The Business Agent is designed to be adaptable, responding to queries and reflecting a company’s voice while managing tasks like addressing FAQs, qualifying leads, and escalating intricate inquiries to human staff when necessary.
Initially, the tool will be offered to businesses at no cost, although paid subscription plans are anticipated soon.
Gleit further elaborated, “We actually want to take actions now. We want it to be able to complete payments, process bookings, and place orders,” indicating a departure from more traditional, rule-based automated responses used by legacy bots.
Alongside the Business Agent, Meta is launching a “Business Agent Platform,” which is aimed at providing businesses with resources to create customized AI agents for diverse operational needs.
This platform connects with numerous non-Meta systems such as Shopify, Zendesk, and Shopee, allowing agents to be used across various channels. It also offers larger companies advanced controls, guardrails, and measurement tools.
Exploring New Business Avenues
Gleit is leading initiatives to broaden the company’s business scope with AI agents, supported by a fresh team called Enterprise Solutions established during a recent restructuring focused on AI development.
This team will deploy groups of engineers to work alongside enterprise clients, an approach adopted by some AI firms like Anthropic, which aims to facilitate AI adoption and customize solutions for better outcomes.
Currently, the team’s focus is on the new business agents while also exploring the creation and sale of AI products for internal business operations.
Gleit mentioned the goal of unifying the various AI solutions Meta has developed, including tools for workflow management, a customer-facing AI support bot, and a recently launched ads-specific “business assistant.” She noted, “The number one thing I hear, especially from small businesses, is ‘I just want to go to one place that can do all the things.’”
She emphasized the importance of modularity and adaptability, stating, “You also need to be willing to evolve, because the technology is moving so quickly.”
Risks Associated with AI Integration
However, Gleit also recognized the potential risks of deeply embedding AI agents into corporate frameworks and granting them the authority to act autonomously.
Meta has faced some setbacks recently with its agents, including a notable incident where hackers managed to convince an AI support chatbot to give access to prominent Instagram accounts.
In response to inquiries about the incident, Gleit explained that the company was still examining the issue but identified a faulty “technical check” within the underlying support system as the source of the problem.
“It wasn’t the agent. The agent actually exposed a technical check that wasn’t working,” she clarified. “There was a separate system and technical check that had a bug, and because people were using the agent, they discovered it.”
