AWS' strategic steps to close the GenAI competitive gap

In the past year, Amazon Web Services (AWS) has been a striking echo of Clayton M. Christensen’s The Innovator’s Dilemma – well-managed companies that watch their competition keenly, listen to their customers, and invest aggressively in new technology but still lose market dominance.

As rival hyperscalers Microsoft and Google dominated industry conversations with their Generative AI (GenAI) products, AWS struggled with a growing competitive gap.

From powerful language models to code-writing assistants to image generators, GenAI flourished in 2023 and found a myriad of applications across a wide range of industries.

AWS saw little of that action but, to Amazon’s credit, the company is taking significant steps to close the competitive chasm. Nowhere was this more obvious than its annual AWS Re:Invent learning conference, held in Las Vegas from 27 November through 1 December 2023, where GenAI took centre stage during keynote addresses.

It made a series of AI-related announcements at the event, all of which will ramp up AI-related innovation in 2024.

Amazon Bedrock

As the name creatively suggests, Bedrock is the foundation model (FM) for GenAI application development. Touted by AWS as the “easiest way to build and scale Large Language Models (LLMs) and other FMs,” the platform makes complex AI development tasks more manageable and user-friendly.

One of its key features is enhanced control. Together with a focus on transparency, this will give developers a clear view of task processing and execution allowing easier troubleshooting and optimisation. Its adaptability is also noteworthy – Bedrock connects to several data sources and systems and is customisable to suit the specific requirements of a task.

Future plans include introducing Guardrails for Bedrock. This feature will help organisations add a layer of protection for GenAI applications to block undesired topics, filter harmful content and redact Personally Identifiable Information (PII), amongst other custom safeguards.

Amazon Q

One of the biggest announcements was the launch of Amazon’s own AI Chatbot for enterprise. AWS says Amazon Q is designed to provide “fast, relevant answers to pressing questions, solve problems, generate content, and take actions using the data and expertise found in information repositories, code, and enterprise systems”.

With more than 40 built-in connectors – including Amazon S3, Microsoft 365, Google Drive, Salesforce, Slack, ServiceNow, Zendesk, Gmail and Atlassian – enterprise users can have tailored conversations, generate content and solve problems.

AWS HealthScribe

The tool uses generative AI to create clinical notes from patient-clinician conversations. Via an API, HealthScribe automatically extracts medical terms, identifies speaker roles, classifies dialogue, and generates detailed preliminary clinical transcripts and notes.

The tool is HIPAA-compliant and can be used by HIPAA-eligible third-party software vendors to build clinical applications that use speech recognition and generative AI to generate preliminary clinical documentation.

AWS HealthScribe is a timely addition to the GenAI ecosystem given the AWS considering that a recent athenahealth survey found that 57% of physicians report that the leading cause of regular burnout is excessive documentation requirements leading to negative patient outcomes.

Partnerships

At AWS, bridging the GenAI competitive gap extends beyond internal development to strategic collaborations with industry leaders such as Nvidia and Siemens. Together with Nvidia, AWS is working to improve the scalability of AI infrastructure. The company has also teamed up with Siemens to integrate Amazon Bedrock into Siemens’ Mendix low-code development platform. This move is expected to help democratise AI.