Amazon is switching on its new Alexa+ chatbot in Australia on Thursday, betting that it will make fewer embarrassing mistakes than rival assistants powered by similar AI. The upgraded system uses a large language model, following a path already taken by Google’s Gemini which has attracted criticism for frequent missteps.
Amazon positions Alexa+ as more reliable on simple facts and less prone to spreading false information, even if it still falls short of perfection. The company frames the rollout as a major step forward, not a finished product.
Behind the upgrade is Amazon’s decision to move its long-running Alexa assistant to the same type of generative AI that now underpins many chatbots. Large language models generate responses based on statistical patterns in data, which makes them sound more natural but also introduces unpredictability.
Amazon highlights that these systems are “non-deterministic”, so the same question can produce different answers and some will be wrong. The company is trying to manage expectations by stressing both the benefits and the built-in limits of the technology.
Alexa+ runs on Echo smart speakers, which have been in Australian homes since 2018 and form the backbone of Amazon’s local voice ecosystem. Amazon says at least 98% of Echo devices sold in Australia are modern enough to receive the upgrade, giving the company an instant installed base for the new AI.
The rollout starts with an “early access” phase so Australian users can test the system and feed back on local performance, accents and use cases. That feedback loop is meant to help tune Alexa+ for Australian conditions, including local information and everyday language.
The Alexa+ launch in Australia illustrates how quickly generative AI is reshaping consumer tech, even as core reliability problems linger. Amazon is betting that its focus on reducing basic errors will give it an edge over Gemini and other assistants that have struggled in public.
The early access approach also indicates that Amazon wants Australians actively involved in shaping how the assistant behaves. Accuracy and trust, not just flashy features, are likely to decide which AI assistant wins the household.
