ITP Techblog

There’s a good reason Google boss Sundar Pichai issued a “code red” to his AI engineers before Christmas - the ChatGPT revolution poses a threat to Google’s vast search engine empire.

Part of Pichai’s reasoning for calling on Google co-founders Larry Page and Sergey Brin to urgently review Google’s AI strategy is the speed at which Microsoft, an investor in ChatGPT creator Open AI, is building the generative AI models into its own products.

Microsoft is OpenAI’s “exclusive cloud provider”, Last week the software giant made the Azure OpenAI Service, which first debuted in 2021, generally available to invited users of its Azure cloud platform. The artificial intelligence models that Microsoft customers will be able to access via Azure include the Generative Pre-Trained Transfomer (GPT) 3.5, the Codex AI that translates natural language to code, and the DALL-E 2 image creation AI.

Both DALL–E 2 and ChatGPT products from OpenAI have caused a blizzard of hype and discussion in recent months as users test drive them online to discover their strengths and limitations - and provide OpenAI with a wealth of data in the process.

Google has its own equivalent of ChatGPT. As the New York Times reports: 

“Google executives hope to reassert their company’s status as a pioneer of A.I. The company aggressively worked on A.I. over the last decade and already has offered to a small number of people a chatbot that could rival ChatGPT, called LaMDA, or Language Model for Dialogue Applications.”

But the company has been slower to ready its chatbot tools for delivery in the cloud, which gives its cloud computing rival Microsoft an advantage. Google has professed a desire to address the potential for large language models to exacerbate the problems of misinformation, bias and hate speech. OpenAI claims to be further ahead in doing so and has clearly given Microsoft the confidence to offer the services to its customers.

The effort to 'democratize AI'

“We formed our partnership with OpenAI around a shared ambition to responsibly advance cutting-edge AI research and democratize AI as a new technology platform,” Satya Nadella, Microsoft’s chairman and CEO, said in a statement yesterday. 

“In this next phase of our partnership, developers and organizations across industries will have access to the best AI infrastructure, models, and toolchain with Azure to build and run their applications.”

In terms of who can access the Azure OpenAI Service, Microsoft said last week: 

“Availability is restricted to customers who meet and adhere to the standards for responsible and ethical AI principles that Microsoft has set and published. Customers are required to apply for access describing their intended use-case or application before they are given access to the service.”

Microsoft’s AI standards for its own products outline various goals covering accountability, transparency, fairness, reliability & safety, privacy & security, and inclusiveness.

Microsoft also has an AI Impact Assessment template it has encouraged users of AI systems to complete to help Microsoft understand how they plan to use Azure OpenAI Service. Examples of general capabilities Microsoft is inviting applicants to outline include: Conversational AI (Such as chatbots, virtual agents, and question answering), search, summarisation, writing assistance or content generation, code-based scenarios (natural language to code, code to code, code to natural language), reason over structured and unstructured data (classification, entity extraction, sentiment analysis), and the DALL-E AI-powered image generator.

New Zealand companies could well be in the front ranks of Microsoft customers testing the water with these tools for everything from computer coding to generating documents, as Azure OpenAI Service is being made available on application worldwide. 

Microsoft products like Power BI already use GPT-3 powered natural language to “generate formulae and expressions”. But the real breakthrough for the likes of ChatGPT could be building them into Microsoft’s more widely used products, including the Office 365 suite and the Bing search engine.

Those applications of AI certainly appear to be on the cards and Microsoft on Monday revealed it will make a large new investment in OpenAI. While it wasn’t naming dollar amounts, speculation in the US suggests it could be as large as US$10 billion, an indication of the potential it sees in OpenAI’s tools to add value to its own products and services.

Read the application form Microsoft is asking customers keen to use Azure OpenAI Service to fill out.