Your colleague that’s happy to do the mundane parts of work: AWS announces frontier AI agents

2 hours ago 1

Advertorial for Amazon Web Services

By Janet Stone

January 13, 2026 — 9.41am

Not long ago, artificial intelligence in the workplace was a helpful assistant, drafting emails, summarising meetings and answering queries on cue. At re:Invent 2025, Amazon Web Services (AWS) unveiled a new class of “frontier agents”, AI technology designed not just to assist, but to autonomously manage complex tasks for hours or even days at a time, freeing up workers to focus on more strategic, impactful, or just more fun work.

Modern professionals focus on high-value creative strategy while autonomous agents manage complex software development.

These agents don’t wait for prompts. They diagnose problems and make decisions within defined guardrails independently, with its goal to complete the task assigned, while checking in with a human to make the final decision. In effect, they operate more like “work assistants” than tools.

Solving complex challenges in minutes, not hours

Among the new releases showcased at re:Invent were three AWS frontier agents. Kiro, an autonomous software development agent, can understand and maintain context across an entire codebase, writing and refining code as a human developer would. DevOps Agent investigates and resolves tech system failures, often stitching together signals across company systems that would otherwise take skilled engineers hours to untangle. And Security Agent performs penetration testing and vulnerability assessments at a scale and speed that was previously impossible.

During testing at Commonwealth Bank of Australia, the DevOps Agent identified the root cause of a complex network and identity-management issue in under 15 minutes, a task that may have taken senior engineers hours. “AWS DevOps Agent thinks and acts like a seasoned DevOps engineer, helping our engineers build a banking infrastructure that’s faster, more resilient, and designed to deliver better experiences for our customers,” says Jason Sandery, head of Cloud Services at Commonwealth Bank. “This isn’t just about faster resolution times, it’s about maintaining the trust our customers put in us.”

To ensure these agents operate securely and responsibly, there are guardrails to provide real-time governance and centralised security controls that protect sensitive data and prevent misuse as the agent is working. These enterprise-grade safeguards ensure compliance and responsible AI deployment.

The agents sit alongside a major expansion of AWS’s Nova AI models, which underpin how these systems perceive, reason and communicate. The new Nova 2 Pro and Nova 2 Lite models balance accuracy and speed for enterprise workloads, while Nova 2 Sonic introduces natural, real-time voice interaction across more than 30 languages.

In practical terms, that opens the door to AI-powered customer service that can hold fluid, conversational dialogue rather than scripted exchanges.

Nova Forge allows organisations to build their own AI models by securely blending their proprietary data with AWS’s Nova models for frontier intelligence. This approach also gives businesses tighter control over accuracy, privacy and intellectual property.

Companies like Reddit, Booking.com, and Sony, are already building their own models with Nova Forge.

Scaling AI across the Australian economy

While AWS agents resolve technical system failures, workers may be able to gain back as much as 23 hours per week for innovation.

AI adoption is accelerating rapidly with half of Australian businesses now consistently using some form of AI, up from 43 per cent in 2024, according to AWS’s latest research. The impact of utilising AI is measurable, with 95 per cent of AI adopters reporting increased revenue, averaging 34 per cent growth. While 86 per cent have experienced productivity gains, saving an estimated 23 hours per week.

Yet a challenge persists. While adoption is widespread, most Australian businesses remain at basic levels. Although 81 per cent of startups use AI throughout their business, with 53 per cent integrating it at their core, 61 per cent of large enterprises remain focused on incremental efficiency gains and only 37 per cent of businesses feel their workforce is prepared for AI.

AWS’s commitment to Australian AI capability is backed by substantial infrastructure investment. The company announced a planned $20 billion investment through 2029 to expand data centres in Australia.

The timing is critical. With 87 per cent of AI adopters believing the technology will fuel their growth in the coming year, and 94 per cent anticipating significant cost savings, averaging 38 per cent, as they integrate AI more deeply into operations. AWS’s infrastructure investment and innovative technology are positioning Australian businesses to capture transformative opportunities.

Agentic AI is already here. The question for Australian organisations isn’t whether AI agents will transform how we work, but how quickly businesses will embrace them to maintain competitive advantage while also delivering better customer experiences.

Read more here.