Off-Prem

SaaS

You begged Microsoft to be reasonable. Instead it made Copilot reason-able with OpenAI GPT-o1

'Magical free' upgrade coincidentally follows M365 price hike


Microsoft has made Think Deeper, OpenAI's GPT-o1 reasoning model, "free and available for all users of Copilot."

This is just two weeks after increasing the cost of Copilot-infused Microsoft 365 by $3 per month and the release of DeepSeek R1, a competitive free and open source AI model from the eponymous China-based AI firm. Both o1 and R1 are so-called chain-of-thought models, in that they produce intermediary stages of output before settling on a final answer, which is supposed to help them iron out hallucinations, errors, and other low-quality stuff. Those intermediary stages emulate a form of artificial reasoning by the LLMs.

Mustafa Suleyman, CEO of Microsoft AI, shared the news of the o1 availability via Redmond's LinkedIn.

"This now gives everyone access to OpenAI’s world-class o1 reasoning model in Copilot, everywhere at no cost," said Suleyman, meaning no additional cost beyond what you're already paying.

"I urge you to give it a try. It’s truly magical."

Suleyman's use of Apple's well-worn marketing term perhaps doesn't capture the mood at Microsoft this week. On Monday, Redmond's stock dipped two percent as investors weighed the impact of DeepSeek's AI models, which even the Windows giant has now rapidly integrated into its portfolio.

Microsoft shares have retreated another six percent since Wednesday (~$416 per share from ~$445) following the IT titan's second quarter earnings report, a drop attributed to lighter than expected Azure revenue growth and revised financial guidance.

Suleyman's loose definition of "free" hasn't gone unnoticed - much like how the tech industry plays fast and loose with the word "open."

"Hmm... within 24 hours of this being announced as being given out 'for free' I have just received an email from Microsoft informing me that my monthly family M365 subscription is going up by 31 percent," said James Crouch, UK director at ElectroNet, responding to Suleyman's post.

"Don't get me wrong, I understand the strategy (especially with the panic around DeepSeek) and I will appreciate having access to o1 in Copilot. But it seems disingenuous to suggest this is being given away for free."

Leaving aside the debate over whether this is truly "free" - and we're just talking about subscription fees here, not the environmental impact or capital expenditures, Suleyman urged users to try Think Deeper.

As mentioned, the the o1 model from Microsoft-bankrolled OpenAI is designed to work through complex tasks in a series of steps in a chain of thought process likened to the way people solve problems.

"Get in-depth advice on how to manage a career change, with detailed breakdowns of educational milestones and options, resources on where to look for roles, strategies for getting in the door and industry trends you absolutely need to know," he said, pointing to various ways to explore the AI model's chain of thought reasoning.

Microsoft employees caught off in this month's job cuts no doubt will be delighted to consult Think Deeper about what comes next. ®

Send us news
25 Comments

We meet the protesters who want to ban Artificial General Intelligence before it even exists

STOP AI warns of doomsday scenario, demands governments pull the plug on advanced models

Microsoft expands Copilot bug bounty targets, adds payouts for even moderate messes

Said bugs 'can have significant implications' – glad to hear that from Redmond

How nice that state-of-the-art LLMs reveal their reasoning ... for miscreants to exploit

Blueprints shared for jail-breaking models that expose their chain-of-thought process

LLM aka Large Legal Mess: Judge wants lawyer fined $15K for using AI slop in filing

Plus: Anthropic rolls out Claude 3.7 Sonnet

After Copilot trial, government staff rated Microsoft's AI less useful than expected

Not all bad news for Redmond as Australian agency also found strong ROI and some unexpected upsides

Despite Wall Street jitters, AI hopefuls keep spending billions on AI infrastructure

Sunk cost fallacy? No, I just need a little more cash for this AGI thing I’ve been working on

UK's new thinking on AI: Unless it's causing serious bother, you can crack on

Plus: Keep calm and plug Anthropic's Claude into public services

Does terrible code drive you mad? Wait until you see what it does to OpenAI's GPT-4o

Model was fine-tuned to write vulnerable software – then suggested enslaving humanity

Why AI benchmarks suck

Anyone remember when Volkswagen rigged its emissions results? Oh...

Under Trump 2.0, Europe's dependence on US clouds back under the spotlight

Technologist Bert Hubert tells The Reg Microsoft Outlook is a huge source of geopolitical risk

Open Source Initiative defends disallowing board candidate after timezone SNAFU

Here's another thing AI can do: Cause conflict around whether it's compatible with the very idea of open source

Hurrah! AI won't destroy developer or DBA jobs

Bureau of Labor Statics warns lawyers and customer service reps to brace for change, says techies will be fine