Beyond the Chatbot: Amazon Connect Horizons 2025 & the Rise of AI Agents

October 2025

Telecoms & Connectivity

Molly Gatford

Senior Research Analyst

It may come as no surprise that AI agents dominated the discussion at this year’s Amazon Connect Horizons, held in London in October 2025.

These systems are designed to interact with their environment, make decisions, and automate tasks — capabilities that are rapidly transforming customer service operations. Unlike traditional chatbots, AI agent systems go further by not only communicating with humans, but also collaborating with other agents in multi-agent systems, expanding the scope of automation even further.

AI agents are already being integrated into customer interactions to handle routine enquiries such as booking confirmations, order status updates, and cancellations. Beyond automating simple processes, they also support human agents, helping deliver more personalised and efficient customer experiences.

The Role of Amazon Connect

Amazon Connect is a cloud-based contact centre platform that enables self-service customer support across voice, chat, and digital channels. It integrates AI-based capabilities throughout its platform, focusing on automation and the personalisation of customer experiences.

Amazon Connect benefits from Amazon’s own experience delivering large-scale customer service for its online marketplace. This allows it to refine its platform using real-world feedback and data. Its AI functionality is natively embedded across all customer interaction channels, with a usage-based pricing model that includes AI capabilities — designed to make it predictable, scalable, and accessible.

Source: Amazon

At the event, one key challenge discussed was not the speed of deployment, but rather the fragmentation of business operations across regions. Misaligned markets and legacy systems often present significant barriers to integration. We believe this reflects a broader industry challenge: fragmentation and poor-quality data will remain major obstacles to large-scale AI agent deployment.

Looking ahead, Amazon Connect is preparing new launches in the AI agent space, aligning with the industry-wide shift towards greater automation. We expect other platforms to follow suit, with enterprises increasingly adopting AI agents as their benefits become clearer over the coming years.

The Future of Customer Experience Platforms with AI Agents

While contact centre solutions such as Amazon Connect are naturally positioned to specialise in customer service due to their agent-first approach, CCaaS providers are not the only ones moving into this space. CPaaS platforms — which have traditionally focused on enabling enterprises to build communication workflows for transactional or authentication use cases — are also recognising the opportunity to enhance customer experiences through AI agents.

This convergence of platforms in the customer experience space has been gathering pace over the last two years, particularly since the rise of generative AI chatbots. We expect this trend to accelerate with the emergence of AI agents, as convergence becomes essential to ensure consistency across the entire customer journey. Enterprises relying on multiple providers for different use cases, such as marketing or customer support, risk creating fragmented ecosystems and inconsistent service quality.

As more players enter the market, the competitive landscape will continue to evolve. The traditional boundaries between CCaaS and CPaaS are expected to fade, giving rise to unified “Customer Experience Platforms” that span marketing, sales, support, and transactional interactions. To stand out, platforms will need to play to their strengths — CCaaS providers will need to extend into marketing and transactional use cases, while CPaaS players must build greater agent-support capabilities.

The Importance of Seamless AI Agent Integration

For enterprises adopting AI agents, the speed and accuracy with which the system identifies and activates the right agent will be key to maintaining a positive customer experience. This must remain consistent across voice and chat channels. Although voice remains more challenging, recent demonstrations show noticeable reductions in latency and increasingly natural, human-like conversations.

Despite these advancements, many enterprises are likely to take a cautious approach to adoption. High levels of automation will be possible for routine, low-risk tasks, but human oversight will still be necessary for complex or high-stakes interactions. Structured workflows will continue to play an important role in ensuring accuracy and reliability.

The Outlook for AI Agent Development

Over the next year, we anticipate an acceleration in the number of startups developing AI agents, increasing the variety of solutions available. Developers must carefully evaluate ROI when designing AI agents — those that automate routine, repetitive tasks will deliver the greatest value, while niche use cases may yield lower returns. Platforms should remain agent-agnostic, allowing enterprises to integrate the best-fit agent for each task.

Although the market remains far from achieving artificial general intelligence (AGI) — where AI can learn and perform any intellectual task a human can — AI agents are fast becoming a reality. We expect automation to advance steadily as these solutions become increasingly self-serviceable, particularly as platforms like Amazon Connect continue to expand their AI agent offerings in the months ahead.

Molly is a Senior Research Analyst at Juniper Research, providing insight, data, and recommendations for established and future markets within the telecommunications sector. Her recent reports have covered topics including Travel eSIM, A2P & Business Messaging, and CPaaS.