Microsoft Copilot 現在免費提供由 OpenAI o1 模型推動的「Voice」及「Think Deeper」
微軟稍早宣布，將在其 Copilot AI 助理中免費、無限次數地提供「Voice」和「Think Deeper」兩項功能，兩項功能皆由 OpenAI 的 o1 模型提供支援。
Copilot 是在去年 10 月獲得了 Voice 功能，允許用戶與 AI 助理進行對話。Think Deeper 則是運用了 o1 模型的推理思考能力，可以回答更加複雜的問題。
微軟在部落格文章中指出，由於運算力的限制，這些剛剛免費化的 Copilot 應用在使用量高的時候，用戶可能會遇到延遲或中斷，而微軟也正在努力擴大伺服器的負載，以盡快讓更多用戶可以無限次使用新功能。不過，真的受不了了的話，Copilot 會員會有優先使用權，在高負載時也能優先使用 Copilot 上最新的 AI 模型，同時也能優先嘗試仍處於實驗階段的新功能。
相關新聞：
Announcing Free, Unlimited Access to Think Deeper and Voice
Microsoft 更新 Copilot AI，未來將能使用語音並看到你瀏覽的網頁
緊貼最新科技資訊、網購優惠，追隨 Yahoo Tech 各大社交平台！
🎉📱 Tech Facebook：https://www.facebook.com/yahootechhk
🎉📱 Tech Instagram：https://www.instagram.com/yahootechhk/
🎉📱 Tech WhatsApp 社群：https://chat.whatsapp.com/Dg3fiiyYf3yG2mgts4Mii8
🎉📱 Tech WhatsApp 頻道：https://whatsapp.com/channel/0029Va91dmR545urVCpQwq2D
🎉📱 Tech Telegram 頻道：https://t.me/yahootechhk
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Forbes
21 minutes ago
- Forbes
It's All Going On: Business And Society In The Next Few Years
To say that there's a lot going on in AI is something of an understatement. Many people less connected to the industry are ignoring its potential at their peril. The technology is due to upend almost everything we know about the supremacy of human capabilities. I'm still sort of pondering a lot of what came out of a recent long essay post on Less Wrong that I started covering last week. The author, who we will call by their handle, L Rudolph L, or simply, for the purposes of this, 'L', brings forth an amazing trove of insights into what life might look like from 2025 through, say, 2029, and as I've said before, these are not vague ambiguous predictions. So let's go through more of how this poster envisions the next part of the AI age playing out. There's a good bit in here about business outcomes. For example: as contenders go for the consumer market, what is OpenAI's moat? L answers this by saying that the company will be 'especially vulnerable, since they rely heavily on consumers, and are also increasingly a product company that competes with products built on their API.' The author's take on the company's likely strategy goes like this: 'Their strategy (internally and to investors, though not publicly) is to be the first to achieve something like a drop-in agentic AI worker and use that to convert their tech lead over open source into >10% of world GDP in revenues. They've raised tens of billions and make billions in revenue from their products anyway, so they can bankroll these efforts just fine.' Meanwhile, the author calls Anthropic a 'jewel of model quality' and a 'Mecca of technical talent', in some great alliteration, while suggesting that XAI and DeepSeek will be the open source leaders. As for winners and losers, here's the take: 'Investors want to see 'real AGI', not just post-scarcity in software. Google DeepMind's maths stuff and xAI's engineering stuff are cool; OpenAI and LLMs are not. Amazon's AWS & physical store is cool, Google Search and Facebook are not.' There's another side point where L talks about 'selling shovel,' or, presumably, the role of infrastructure, and points to Nvidia's success as ongoing in some way, shape or form. We've seen Nvidia stock slump over the last few weeks, but the jury is still out on what that company will look like a couple of years from now. Noting that the progress of AI professionalism might be slowed by human bureaucracy, L Rudolph L nevertheless predicts that AI will learn to perform the role of a physician, and an attorney. Eventually, the human in the loop that the author posits will be mainly relegated to tedious small tasks and data entry, as well as anything that's too sensitive for AI to handle upfront, for instance, certain kinds of verification and authorization tasks. Consider, if you will, the prophetic visions of call centers: 'A surprisingly retro success is call centers of humans who are ready to jump in and put an AI agent back on task, or where AI agents can offload work chunks that are heavy on trust/authentication (like confirming a transaction) or on button-clicking UI complexity (like lots of poor legacy software), to human crowdworkers who click the buttons for them, while the AI does the knowledge/intelligence-intensive parts of the job on its own.' On a sidenote, L also imagines the Nobel prize might be contested between a human and an AI. I liked this part (italics mine): 'Credit for the Nobel Prize is the subject of much discussion, but eventually (in 2030) ends up split between Demis Hassabis, one of the physicists who was most involved, and the most important AI system. Everyone has Opinions.' Theorizing that the Communist Chinese Party unveiled its 15th five year plan in 2026, L also talks about the rivalry between the two major empires – the middle kingdom in Asia, and America in the west. The author also notes that companies like Huawei may be close on Nvidia's heels. The American industry and government, L notes, will be likely focusing on strategies to contain China and prevent it from getting key technologies to dominate the global market and pursue espionage goals. 'There's increasing demand for really air-tight software from a US defense establishment that is obsessed with cyber advantage over especially China, but also Russia, Iran, and North Korea,' L writes. There's even a part where L suggests that human populations will decline because people will be choosing to spend time with their robotic companions instead of other flesh and blood humans, whereby the old manual system of procreation may take a backseat. Besides a trend toward human-AI romances, L also goes into some of how the rank and file may react to these changes. The author talks about 'ChatGPTese' as the noticeable product of the LLM to human audiences. Here's the take on what's likely to happen with AI fakers calling humans on the phone or whatever interface is predominant: 'AI scam calls with deepfaked audio and video start being a nuisance by mid 2025 but are mostly reined in by a series of security measures pushed by platforms (and by regulation in the EU), people creating new trust protocols with each other ('what's our secret passphrase?'), increased ID verification features, and growing social distrust towards any evidence that's only digital.' Basically, L predicts, we will learn to live with normalized AI and what that looks like, as a world with intense automation emerges. Most of us know this intuitively, are preparing for it now. But I would think that for most people, it's extremely hard to envision just how this will work. In any case, do yourself a favor, read the entire essay, and think about all of these predictions as a whole, to try to read the tea leaves on the rest of the 21st century.
Business Wire
43 minutes ago
- Business Wire
HubSpot Launches First CRM Deep Research Connector With ChatGPT
BOSTON--(BUSINESS WIRE)--More than 250,000 businesses rely on HubSpot as their single source of truth for customer data across marketing, sales, and service. This complete view of the customer journey gives our customers an edge, especially in an era where AI is only as powerful as the data behind it. Today, we're excited to announce that HubSpot is the first CRM to launch a deep research connector with ChatGPT. With over 75% of HubSpot customers already using ChatGPT*, we're making it easy for them to apply powerful, doctorate-level research and analysis to their own customer data and context–and to put those business insights to work. This is game-changing for go-to-market teams. Within ChatGPT, for example: Marketers can ask 'find my highest-converting cohorts from recent contacts and create a tailored nurture sequence to boost engagement,' then use the insights to launch an automated workflow in HubSpot. Sales teams can find new opportunities by asking, 'segment my target companies by annual revenue, industry, and technology stack. Based on that, identify the top opportunities for enterprise expansion,' then bring them back to HubSpot for prospecting. Customer success teams can say, 'identify inactive companies with growth potential and generate targeted plays to re-engage and revive pipeline,' then take those actions in HubSpot to drive retention. Support teams can say, 'analyze seasonal patterns in ticket volume by category to forecast support team staffing needs for the upcoming quarter,' and activate Breeze Customer Agent in HubSpot to handle spikes in support tickets. 'Launching the HubSpot deep research connector means businesses and their employees get faster, better insights because ChatGPT has more context. We're thrilled to work together to bring powerful AI to many of today's most important workflows.' - Nate Gonzalez, Head of Business Products at OpenAI. 'The HubSpot connector is like having an extra analyst on the team, empowering sales reps to identify risks, opportunities, and next best actions,' said Colin Johnson, Senior Manager, CRM at Youth Enrichment Brands. 'For a non-technical user, the fact that it's easy to use and talks directly to my data is huge.' 'We're building tools that help businesses lead through the AI shift, not just adapt to it,' said Karen Ng, SVP of Product and Partnerships at HubSpot. 'By connecting HubSpot CRM data directly to ChatGPT, even small teams without time or data resources can run deep analysis and take action on those insights — fueling better outcomes across marketing, sales, and service.' Easy to use and easy to trust HubSpot customers who have admin controls can enable the connector for their organization by going to ChatGPT and turning on the HubSpot deep research connector function, selecting HubSpot as a data source, and authenticating their account. From there, any user in the organization can toggle it on, sign in, and start asking questions. In addition to being easy to use, the HubSpot deep research connector is also easy to trust. We built it to ensure users only see the CRM data they're allowed to access in HubSpot. For example, individual sales reps will only see pipeline data for deals they own or manage. With the HubSpot deep research connector, customer data is not used for AI training in ChatGPT. Availability The HubSpot deep research connector will automatically be available to all HubSpot customers across all tiers with a paid ChatGPT plan. (EU: Team, Enterprise, and Edu; all other regions: Team, Enterprise, Pro, Plus, and Edu). All available languages can be found here. *HubSpot 2025 Q1 AI customer sentiment survey About HubSpot HubSpot (NYSE: HUBS) is the leading AI-powered customer platform for growing businesses. The platform includes engagement hubs for marketing, sales, and customer service, a connected Smart CRM, and an ecosystem of over 1,800 integrations—all built on a unified data foundation that powers HubSpot's AI and enables smarter, faster, more personalized customer experiences. More than 250,000 customers across 135+ countries use HubSpot to unify their data, align their teams, and grow better. HubSpot is headquartered in Cambridge, Massachusetts. Learn more at
Yahoo
43 minutes ago
- Yahoo
HubSpot Launches First CRM Deep Research Connector With ChatGPT
Over 250,000 businesses trust HubSpot as their source of truth for customer data Customers can now bring their customer context into the HubSpot deep research connector and take action on insights The new connector is the latest step in HubSpot's mission to democratize advanced technology for SMBs BOSTON, June 04, 2025--(BUSINESS WIRE)--More than 250,000 businesses rely on HubSpot as their single source of truth for customer data across marketing, sales, and service. This complete view of the customer journey gives our customers an edge, especially in an era where AI is only as powerful as the data behind it. Today, we're excited to announce that HubSpot is the first CRM to launch a deep research connector with ChatGPT. With over 75% of HubSpot customers already using ChatGPT*, we're making it easy for them to apply powerful, doctorate-level research and analysis to their own customer data and context–and to put those business insights to work. This is game-changing for go-to-market teams. Within ChatGPT, for example: Marketers can ask "find my highest-converting cohorts from recent contacts and create a tailored nurture sequence to boost engagement," then use the insights to launch an automated workflow in HubSpot. Sales teams can find new opportunities by asking, "segment my target companies by annual revenue, industry, and technology stack. Based on that, identify the top opportunities for enterprise expansion," then bring them back to HubSpot for prospecting. Customer success teams can say, "identify inactive companies with growth potential and generate targeted plays to re-engage and revive pipeline," then take those actions in HubSpot to drive retention. Support teams can say, "analyze seasonal patterns in ticket volume by category to forecast support team staffing needs for the upcoming quarter," and activate Breeze Customer Agent in HubSpot to handle spikes in support tickets. "Launching the HubSpot deep research connector means businesses and their employees get faster, better insights because ChatGPT has more context. We're thrilled to work together to bring powerful AI to many of today's most important workflows." - Nate Gonzalez, Head of Business Products at OpenAI. "The HubSpot connector is like having an extra analyst on the team, empowering sales reps to identify risks, opportunities, and next best actions," said Colin Johnson, Senior Manager, CRM at Youth Enrichment Brands. "For a non-technical user, the fact that it's easy to use and talks directly to my data is huge." "We're building tools that help businesses lead through the AI shift, not just adapt to it," said Karen Ng, SVP of Product and Partnerships at HubSpot. "By connecting HubSpot CRM data directly to ChatGPT, even small teams without time or data resources can run deep analysis and take action on those insights — fueling better outcomes across marketing, sales, and service." Easy to use and easy to trust HubSpot customers who have admin controls can enable the connector for their organization by going to ChatGPT and turning on the HubSpot deep research connector function, selecting HubSpot as a data source, and authenticating their account. From there, any user in the organization can toggle it on, sign in, and start asking questions. In addition to being easy to use, the HubSpot deep research connector is also easy to trust. We built it to ensure users only see the CRM data they're allowed to access in HubSpot. For example, individual sales reps will only see pipeline data for deals they own or manage. With the HubSpot deep research connector, customer data is not used for AI training in ChatGPT. Availability The HubSpot deep research connector will automatically be available to all HubSpot customers across all tiers with a paid ChatGPT plan. (EU: Team, Enterprise, and Edu; all other regions: Team, Enterprise, Pro, Plus, and Edu). All available languages can be found here. *HubSpot 2025 Q1 AI customer sentiment survey About HubSpot HubSpot (NYSE: HUBS) is the leading AI-powered customer platform for growing businesses. The platform includes engagement hubs for marketing, sales, and customer service, a connected Smart CRM, and an ecosystem of over 1,800 integrations—all built on a unified data foundation that powers HubSpot's AI and enables smarter, faster, more personalized customer experiences. More than 250,000 customers across 135+ countries use HubSpot to unify their data, align their teams, and grow better. HubSpot is headquartered in Cambridge, Massachusetts. Learn more at View source version on Contacts Kim Soward845-741-8201ksoward@ Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data
