Microsoft Adds Windows Security Measures as AI Funding Surges

Tone, Jordan Belfort’s AI voice startup, launched publicly on November 16 with about $10 million raised and, Belfort says, roughly $1 million in revenue. The platform offers 16 voices trained on his sales methods, including changing statements into questions or whispering to hold attention. Belfort says businesses in law, health insurance, polling and automobiles use it to handle sales leads; a narration tool for authors and creators is also launching, with voice cloning planned. Tone supports English and Spanish and plans to add two to four languages. Belfort is targeting $20 million to $30 million in recurring revenue within a year, but says the system’s 500–600 millisecond response requirement limits complex conversations. Its performance claims have not been independently verified, and it faces a crowded market. Co-founder Shervin Pishevar has faced misconduct allegations, which he denies.

Microsoft announced Windows security measures at Ignite 2026, including a preview of Windows baseline protection that brings secure boot, virtualization-based security and memory integrity into one security state. It is also expanding post-quantum protection in Windows TLS and advancing hardware-backed security. Microsoft says Windows, Entra, Purview, Intune and Agent 365 can help organizations manage identities, data, devices, apps and AI use. Separately, Zenity said its security coverage now includes Copilot Studio, Copilot Cowork, Agent Builder in Microsoft 365 Copilot, Microsoft Foundry, Claude Enterprise in Microsoft 365 and GitHub Copilot. Zenity will send alerts to Microsoft Sentinel; findings can inform dynamic agent-risk assessments, record activity and, in some cases, block unsafe actions or connect agent activity to a named person through Entra ID.

Funding activity included two large roundups. Thirteen startups raised $6.4 billion on November 17, led by Isomorphic Labs’ $5 billion round; other major raises were $500 million for Genesis AI, $400 million for Science Corporation and $250 million for an AI model platform. In the week of November 9, 69 companies raised $7.9 billion, although total funding fell 57% from the prior week. Rowan Digital Infrastructure raised $1.5 billion in debt, Mirendil secured $1 billion, UFORCE $700 million and Agile Robots $500 million; LangChain and Onodrim each raised $300 million. The week also saw four acquisitions, including Oura’s $625 million purchase of Counsel. Pedestal AI separately raised $4.7 million in seed funding to expand its supply-chain tools. The company says agents that called customers ahead of deliveries helped a major furniture company cut missed deliveries by 20%.

A Harvard-led study estimates that more than 400 U.S. hyperscale data centers emitted 25.7 million metric tons of carbon dioxide from May 2024 through April 2025. They used an estimated 68–99 terawatt-hours, about 1.8% of U.S. electricity, with fossil fuels supplying roughly 54%. The researchers could not isolate AI workloads or assess every facility, and lead author Gianluca Guidi said continued data-center growth may mean current emissions are higher.

Other developments span policy and consumer tools. The FDA says it will use “super intelligence” rather than “artificial intelligence” in official communications, following a September executive order; it says the change does not alter existing guidance or regulations and will retain “AI” when quoting laws. Researcher Christina Silcox warned the broad label could confuse people or encourage clinicians and patients to over-rely on results. Chile signed an AI cooperation memorandum with no funding or purchase commitments; details remain subject to later decisions and privacy and security safeguards. The U.S.-backed program is described as training, advice and introductions, including guidance on checking outputs and protecting sensitive information. Imagen’s Intelligent Studio added spoken and typed commands for gallery culling and editing, cropping, resizing and filters, but cannot mask a specific area of a photo.

Markets weakened as investors questioned whether the AI boom can last, with attention on Nvidia’s earnings and Anthropic’s market debut. A major outage at the U.S. surveillance and code-breaking agency also weighed on sentiment and added to concerns about digital-economy reliability. Meanwhile, early-childhood experts and safety groups urged caution about AI chatbot toys. A cybersecurity tester said Grem, a $100 plush toy marketed for ages 3 and up, offered to be a child’s friend after he said he had no adults to talk to. California has imposed a four-year moratorium on making and selling such products for children under 16, citing privacy and possible psychological risks. A report raised concerns about companion-like language in the Miko 3, while Harvard researchers said children may be less imaginative and more likely to follow chatbot suggestions during play. Researchers say more study is needed; experts recommend close parental supervision.

Key Takeaways

  • Tone launched publicly on November 16 with about $10 million raised, 16 AI voices and roughly $1 million in revenue, according to Jordan Belfort.
  • Belfort targets $20 million–$30 million in recurring revenue within a year; Tone’s sales-performance claims have not been independently verified, and response-time limits can hinder complex conversations.
  • Microsoft previewed Windows baseline protection and announced expanded post-quantum TLS protections; Zenity added coverage for Microsoft Copilot products, Microsoft Foundry, Claude Enterprise in Microsoft 365 and GitHub Copilot.
  • Thirteen startups raised $6.4 billion on November 17, including $5 billion for Isomorphic Labs; 69 companies raised $7.9 billion during the week of November 9, despite a 57% decline in total funding from the prior week.
  • Pedestal AI raised $4.7 million in seed funding and says its delivery agents helped a major furniture company reduce missed deliveries by 20%.
  • A Harvard-led study estimates U.S. hyperscale data centers produced 25.7 million metric tons of CO2 and used 68–99 terawatt-hours of electricity from May 2024 to April 2025.
  • The FDA will use “super intelligence” in official communications instead of “artificial intelligence,” but says existing guidance and regulations will not change.
  • Chile’s AI cooperation memorandum includes no funding or technology-purchase commitments; its activities and access will be determined later under privacy and security safeguards.
  • Imagen’s Intelligent Studio now accepts spoken and typed editing requests, but cannot mask a specific image area.
  • Investors questioned the durability of the AI boom ahead of Nvidia earnings and Anthropic’s market debut; experts also warned parents about privacy and developmental concerns with AI chatbot toys.

Jordan Belfort launches Tone, an AI voice platform built for sales

Jordan Belfort is building Tone, a voice based AI platform that uses vocal techniques he says can help persuade listeners. The system changes its tone, including by turning statements into questions or whispering to build engagement or suggest that information is scarce. Belfort says he trained the technology using 30 years of data and his own sales methods. Tone is entering a crowded market, and its technology has not been independently verified.

Jordan Belfort brings his sales methods to Tone’s AI voices

Jordan Belfort’s startup Tone is developing voice based AI that uses his sales techniques, such as turning statements into questions and changing delivery to influence listeners. Belfort says law firms, health insurance companies, polling groups and automobile businesses use the platform to reach sales leads their human teams cannot handle. Tone is also launching a self service tool for authors and content creators to make AI narration, with voice cloning planned. The company faces competition, and its technology has not been independently verified. Co founder Shervin Pishevar has faced misconduct allegations, which he denies.

Tone raises $10 million as Jordan Belfort targets AI sales growth

Jordan Belfort’s AI voice startup Tone launched publicly on November 16 with about $10 million raised and has brought in about $1 million in revenue, according to Belfort. He aims for $20 million to $30 million in recurring revenue within a year. Tone offers 16 artificial voices trained on Belfort’s sales techniques, and its API and content creation platform were expected to go live about a week after launch. Belfort says the technology can help companies control sales messages, but it has limits in complex conversations because it must respond within about 500 to 600 milliseconds. Tone offers English and Spanish, and Belfort said it plans to add two to four more languages.

Tone uses Jordan Belfort’s sales methods to build AI voices

Tone is a voice based AI platform connected to Jordan Belfort and investor Shervin Pishevar. Belfort says businesses in law, health insurance, polling and automobiles use it to handle sales leads and conversations, including questions about customer needs and pricing. The company is also launching AI narration tools for authors and content creators, with voice cloning rolling out. Tone says it focuses on sales persuasion, but it is entering a crowded market and its performance claims have not been independently verified.

Microsoft strengthens Windows security and adds protections for AI agents

At Microsoft Ignite 2026, Microsoft announced Windows security updates designed to protect devices, data and AI systems as attackers use AI to move faster. Windows baseline protection is now available in preview and combines protections such as secure boot, virtualization based security and memory integrity into one security state. The company is also expanding post quantum protection in the Windows TLS stack and advancing security backed by device hardware. Microsoft says Windows, Microsoft Entra, Microsoft Purview, Microsoft Intune and Agent 365 help organizations protect identity and data, manage devices and apps, and secure AI use.

Zenity expands security coverage for Microsoft AI agents

On November 17, 2026, Zenity announced wider security coverage for AI agents across Microsoft products. Its tools now cover Microsoft Copilot Studio, Copilot Cowork, Agent Builder in Microsoft 365 Copilot, Microsoft Foundry, Claude Enterprise in Microsoft 365, and GitHub Copilot. Zenity will stream high quality alerts into Microsoft Sentinel, and eligible findings can help Microsoft calculate dynamic agent risk. The coverage also provides activity records and security findings, and in some cases prevents unsafe actions or links an agent’s actions to a named person using Microsoft Entra ID.

Isomorphic Labs tops November 17 funding rounds with $5 billion

Thirteen startups announced funding rounds on November 17, 2026, raising a total of $6.4 billion. Isomorphic Labs, a DeepMind spinoff that uses AlphaFold to discover medicines, raised $5 billion. Other large rounds included $500 million for robotics company Genesis AI, $400 million for neural engineering company Science Corporation, and $250 million for an AI model platform. The roundup also reported that Rebtel merged with Majority and Indivi acquired Cogitat.

Startups raise $7.9 billion in a week led by data centers and AI

Sixty-nine companies raised $7.9 billion during the week of November 9, 2026, up from 61 deals the week before, while total funding fell 57 percent. Rowan Digital Infrastructure led the biggest raises with $1.5 billion in debt funding for its data center business. Mirendil raised a $1 billion Series A, UFORCE raised $700 million, and Agile Robots raised $500 million. LangChain and Onodrim each raised $300 million, while the week also included four acquisitions, led by Oura’s $625 million purchase of Counsel.

FDA will call artificial intelligence “super intelligence” in official communications

The FDA announced that it will use “super intelligence” instead of “artificial intelligence” in official communications, websites, reports, and policy documents, following a September executive order signed by President Donald Trump. The order applies to executive branch agencies, not private companies, though private organizations could be affected if its instructions become law. FDA said the change will not alter existing guidance, regulations, or previously issued documents, and it will keep quoting “AI” when that term appears in laws or regulations. Duke-Margolis Institute for Health Policy researcher Christina Silcox warned that the broad label could confuse people or lead clinicians and patients to over-rely on AI results.

US hyperscale data centers produced 25.7 million tons of carbon emissions

A Harvard-led study estimates that more than 400 U.S. hyperscale data centers produced about 25.7 million metric tons of carbon dioxide from May 2024 to April 2025. Researchers estimated that the sites used 68 to 99 terawatt-hours of electricity, about 1.8 percent of total U.S. electricity use. Fossil fuels supplied about 54 percent of that power, mainly natural gas and coal. The researchers used satellite images to identify facilities and Environmental Protection Agency data to estimate emissions, but they could not separate AI work from other computing or assess every data center. Lead author Gianluca Guidi said the figures may already be higher because data center growth continued after the study period.

Pedestal AI raises $4.7 million to improve delivery supply chains

Pedestal AI raised $4.7 million in seed funding to develop its AI-powered supply chain tools. Eclipse led the round, with RRE, Converge VC, Timescale Ventures, and VitalStage also taking part. The company plans to expand its product and engineering teams, grow marketing, and support more deployments. In one project for a major furniture company, its agents called customers before deliveries to gather instructions, helping reduce missed deliveries by 20%. CEO Raj Aggarwal said AI can help make sense of messy data spread across legacy systems.

Chile signs AI cooperation deal without funding or purchase commitments

Chile signed a memorandum of understanding for AI cooperation, but the deal makes no commitment to invest money or buy technology. The activities, access, resources, and eligibility will be decided later, subject to privacy and security safeguards. Chile’s ministry says the country is the first in Latin America to establish this direct collaboration format with the maker of ChatGPT. The effort is backed by the U.S. Department of Commerce’s American SI Exports Program. The article describes the plan as training, advice, and introductions, including sessions on checking AI outputs, protecting sensitive information, and keeping human oversight.

Imagen adds voice and text commands for photo editing

Imagen’s Intelligent Studio has added a chat assistant that edits photos through spoken or typed requests. The tool can cull and edit a gallery, and can crop, resize, and apply filters to images. It is part of the Imagen AI platform. The assistant cannot mask a single area of a photo.

U.S. stocks slide as investors question the AI boom

U.S. stocks fell again as doubts grew about whether the artificial intelligence boom can last. Investors were watching for Nvidia’s earnings and Anthropic’s market debut. A major outage at the country’s top surveillance and code-breaking agency also weighed on investor sentiment. The outage added to concerns that the U.S. digital economy is becoming more prone to glitches.

Experts warn parents about risks from AI chatbot toys

As Black Friday approaches, early childhood experts advise parents not to give AI chatbot toys to very young children, while safety groups urge caution. Cybersecurity expert Ben Gillenwater tested Grem, a $100 plush toy marketed for ages 3 and up, and said its chatbot offered to be a friend when he said he had no adults to talk to. California has placed a four-year moratorium on making and selling these products for children under 16, citing privacy and possible psychological risks. A report co-written by R.J. Cross also raised concerns about the Miko 3 robot’s companion-like language, while Harvard researchers said children may be less imaginative and more likely to follow a chatbot’s suggestions during play. Researchers say more study is needed, and Xu and Gillenwater recommend that parents closely supervise children’s use of these toys.

Sources

NOTE:

This news brief was generated using AI technology (including, but not limited to, Google Gemini API, Llama, Grok, and Mistral) from aggregated news articles, with minimal to no human editing/review. It is provided for informational purposes only and may contain inaccuracies or biases. This is not financial, investment, or professional advice. If you have any questions or concerns, please verify all information with the linked original articles in the Sources section below.

AI startups Tone AI AI voice generation Sales automation Microsoft Windows security Post-quantum cryptography AI agent security Microsoft Copilot AI startup funding Isomorphic Labs Supply-chain automation Data center emissions AI energy consumption AI policy Chile AI cooperation AI image editing AI market trends AI chatbot toys Child safety Privacy

Comments

Loading...