Microsoft, Google rebuild their worlds around AI
Major product announcements this week from Microsoft and Google show how determined both tech giants are to build generative AI into the heart of their computing worlds.
Why it matters: Tech's giants see AI both a new competitive race and a chance to breathe new life into the central franchises that drive their businesses — like Microsoft Windows and Google Search.
Driving the news: Microsoft has been adding AI assistants to apps and products all year. But at a fall event Thursday, the company announced a new plan to unify all of them into a single Microsoft Copilot that will use knowledge from across apps and data sources and have both work and consumer uses.
- In a new update to Windows 11 available next week, Microsoft is adding its Windows Copilot, bringing Dall-E 3 to Bing and adding AI features to apps such as Paint, Photos and its screenshot tool.
- Windows will also using AI to simplify a range of tasks such as backing up a computer and moving to a new machine.
Google, meanwhile, has been on its own race to incorporate AI across its products.
- This week, Google announced improvements to its Bard chatbot designed to help it offer more reliable answers.
- That's crucial for the company, which has built its whole business around providing trustworthy information and knows it can't use conversational AI to answer users' questions unless it tames the technology's tendency to "hallucinate" (make stuff up).
- The changes to Bard also allow users to connect the chatbot to their troves of Google data, including their Gmail archives and Google Docs files.
On Thursday, Google also showed off a bunch of new features that tap AI to make video editing and publishing easier for YouTube creators.
- One of these, Aloud, lets moviemakers dub their work into another language with just a single click within YouTube Studio.
- Those announcements follow a July reorganization of the team that builds Google Assistant, first reported by Axios, designed to remake the tool for the era of generative AI.
The big picture: For many years, choosing a computing platform meant picking between desktop operating systems (Mac or PC), browsers or mobile platforms (iOS or Android).
- AI is changing that equation, providing an entirely new set of comparison points for users to consider in picking their digital tools.
- By helping users accomplish tasks with plain-English questions and commands, AI is giving Microsoft and Google a chance to fundamentally reimagine what their core products can do for users.
State of play: For Microsoft in general, and Windows in particular, it's particularly advantageous that the generative revolution is occurring now, rather than a decade or two earlier.
- In that time, Microsoft has quietly built up a range of technology assets beyond its desktop franchises — from its OpenAI partnership to its cloud computing prowess with Azure.
- Perhaps just as important, Microsoft is finally far enough removed from its epic antitrust battles that it is willing — and seemingly able — to integrate significant new capabilities into Windows without raising regulatory alarms.
Be smart: Integrating Microsoft's various Copilots into a more unified meta-assistant will take time. Microsoft's Thursday announcements declared its long-term intent while outlining the initial steps it is taking with Windows 11.
- The fragmentation of the technology so far — with separate AI features in separate products —is a result of how fast it has been evolving, Aaron Woodman, Microsoft's vice president of Windows marketing, told Axios.
- "The next logical step is convergence," Woodman said.
Between the lines: For Woodman, the energy inside Microsoft right now has a "renewed hunger" that reminds him of college.
- That's very different, he says, from most of his 26 years at the software giant, working on projects like search and mobile phones — where the company seemed to be waiting for the technology to evolve before choosing its ground.
Missing in action in this competition, for now, is Apple.
- The iPhone's market dominance means that OpenAI, Microsoft, Google and others will surely bring their tools onto Apple's turf. But Apple has been extremely cautious about adding features to its own systems and services based on the latest generative AI technology.
- Dictation and autocomplete did get a boost from the underlying transformer technology in iOS 17.
- And the company is said to be internally testing its own GPT-like engine, per Bloomberg.
"We view AI as huge and will continue weaving it into our products on a very thoughtful basis," Apple CEO Cook said during an earnings conference call last month.
- While talking about its potential — and acknowledging Apple is investing heavily in research — Cook added a note of caution. "I do think it's very important to be deliberative and thoughtful," he said. "There's a number of issues that need to be sorted."
Editor's note: This story has been updated to clarify that Windows Copilot is arriving via a new update to Windows 11, not a whole new edition of the operating system.