Meta Secures $15B for AI, Backs New Robots.txt Signal with Google & Anthropic

Key Takeaways

• <ul><li>Meta raised $15 billion in bonds to fund AI infrastructure, with $135B-$145B in expected capital spending this year</li><li>Google, Anthropic, and OpenAI back a new __usage_ai-training robots.txt signal to let publishers opt out of AI training</li><li>OpenAI launched Daybreak Red in AWS GovCloud, offering GPT-5.6 Cyber for government cybersecurity testing</li><li>Sterling Infrastructure has over $7B in backlog tied to data center projects as Texas data center investment shifts to other states</li><li>3M expanded its partnership with AWS to scale optical connectivity for AI data centers, projecting 500+ jobs and $500M economic contribution</li><li>Meta acquired Conductor, an AI coding startup using agents from Anthropic and OpenAI, though its cloud service shuts down November 20</li><li>10x Genomics launched Sentira Single Cell, a cloud AI app built on Anthropic's Claude for single cell analysis</li><li>Broadcom released TrueSource Spring Enterprise and Arcot Scam Defense tools to integrate security into developer workflows</li><li>Innodisk is supplying eSSDs and DDR5 memory modules for AI data center servers</li><li>Copado acquired Valo.ai to give Salesforce teams a prioritized view of security risks across connected applications</li></ul>

3 US AI Infrastructure Stocks Beyond Texas Data Center Growth

Texas has paused its data center boom, shifting capital to other states. Three US companies stand to benefit from this relocation. Sterling Infrastructure has over US$7b in backlog weighted toward data center projects. Amphenol provides connectors and cables for AI infrastructure with a US$197.1b market cap. nVent Electric supplies cooling and power equipment, with new liquid cooling facilities planned in Minnesota.

10x Genomics Launches Sentira AI Cloud Tool Before Earnings

10x Genomics launched Sentira Single Cell, a cloud-based AI app built on Anthropic's Claude. It lets researchers run natural language single cell analysis on Chromium data. The launch shifts more workflow to the cloud and increases recurring software dependencies. The company is also appointing new board members Ansuman Satpathy and Joseph Gilliam. Q3 earnings are expected on Thursday as a key test of progress.

3M Expands AI Data Center Optics With AWS Partnership

3M and Amazon Web Services expanded their collaboration to scale 3M's Expanded Beam Optical platform for data centers. The initiative aims to support next-generation optical connectivity for AI infrastructure. It is expected to create over 500 jobs and contribute nearly US$500 million to the economy. 3M has been building a roughly US$600m data center focused activity using its materials expertise. Investors will watch for revenue progress on upcoming earnings calls.

Broadcom Unveils New AI Security and Fraud Defense Tools

Broadcom announced new security offerings including TrueSource Spring Enterprise and Arcot Scam Defense. TrueSource Spring Enterprise focuses on automated vulnerability fixes for software supply chains. Arcot Scam Defense is designed to identify and block real-time payment scams. Broadcom also partnered with JFrog on universal zero-touch remediation. These tools aim to integrate security into everyday developer workflows alongside its chip business.

3 AI Infrastructure Stocks in Oil and Data Center Buildout

Aramco is investing billions into AI and expanding infrastructure amid rising Turkish inflation and tight European gas storage. Three stocks are exposed to these trends. Arrow Electronics supplies hardware for AI data centers with US$25.4b in sales. Sanmina builds electronics and racks for AI data centers with a US$8.9b market value. Jabil designs server platforms for AI data centers with a US$28.1b market value.

Google supports new robots.txt signal for AI training control

Google will support a unified robots.txt signal that lets website owners control how their content is used for AI training. The signal is called __usage_ai-training and can be added to existing robots.txt files. Google plans to support this control in 2027 along with other AI providers. The proposal was created by Kevin Kelley at Anthropic and is being considered by the IETF. Google will read the token and treat sites without it as denying Google-Extended if they do not reference it.

New policy token could help publishers opt out of AI training

A proposed policy token aims to make it simpler for publishers to opt out of AI training. OpenAI, Anthropic and Google have backed the proposal, but it is not yet a standard. Publishers are advised to use data protection policies and technical measures in the meantime. The European Union has also proposed a law requiring explicit consent for AI training. It is still unclear whether either the token or the law will become widely adopted.

Parents upset over school AI career tool for 6-year-olds

Parents are angry after an unnamed US school used AI to show 6-year-olds what they might look like in future careers. The system analyzes photos of the children as adults. Parents say it is an invasion of privacy and that they did not agree to it. The school has not commented on the controversy. The incident has sparked a wider debate about AI use in education.

Meta raises $15 billion in bond sale to fund AI expansion

Meta Platforms announced a $15 billion US dollar bond offering in five parts to fund AI infrastructure investments. The bonds include $3.25 billion due in 2031 at 5.85 percent and $3.25 billion due in 2033 at 6.25 percent. This comes days after Meta expanded plans for a possible $97.5 billion equity raise. Meta expects capital spending of $135 billion to $145 billion this year, largely for AI. Moody's rated the new debt Aa3.

Innodisk expands AI data center push with eSSD and DDR5

Innodisk is deepening its AI data center strategy by supplying key components for AI infrastructure. The company is providing enterprise solid-state drives (eSSDs) and AI server memory modules including DDR5 products. The move was reported on November 4, 2026. Innodisk is described as an industrial memory module maker. The announcement highlights its focus on critical hardware for AI data centers.

Astrom brings AI pilot training framework to EATS summit

Astrom Training Solutions is presenting its GUN Model at the European Aviation Training Summit in Cascais, Portugal. The AI-assisted framework connects training, testing, remedial training and curriculum adjustment in a continuous cycle. It supports task-based and competency-based grading through a common evidence structure. Instructors can also use it for concordance training to compare scoring consistency. The framework tracks individual pilot performance against fleet requirements and development targets.

Daybreak Red cybersecurity model now available in AWS GovCloud

OpenAI has made Daybreak Red available in AWS GovCloud (US-East) for approved government customers. The service provides access to GPT-5.6 Cyber, a purpose-trained cybersecurity model for vulnerability research and security testing. Government teams can use it to investigate software weaknesses and develop mitigations through Amazon Bedrock. OpenAI states that inference data is not used to train models and customers do not share their data with OpenAI. Access requires both OpenAI Daybreak Red approval and AWS provisioning.

Copado acquires Valo.ai to strengthen Salesforce security visibility

Copado has acquired Valo.ai to help Salesforce administrators monitor production security risks. Valo.ai was founded by former leaders from Salesforce Shield. Its platform gives teams a prioritized view of connected applications, permissions, dormant integrations and unused accounts. Each user and integration receives a risk score to help identify unnecessary access. Valo will remain available as a standalone product while Copado plans phased integration into development and release workflows.

Meta acquires AI coding startup Conductor

Meta has acquired Conductor, an AI coding startup founded in 2024 that lets engineers deploy agent teams on projects. The agents come from external labs like Anthropic and OpenAI rather than being built in-house. Conductor's platform works in the cloud, locally or via a user's phone. The company's core features will continue functioning until early next year. However, the cloud service will shut down on November 20, after which user cloud data will be deleted.

Bites raises $40 million to expand AI restaurant training

Bites has raised $40 million to grow its AI training technology for restaurant employees. The platform distributes training across shifts and locations through channels like WhatsApp, SMS and email. Its Ask AI feature lets employees look up information using approved materials during shifts. Bites is also connecting training data with POS, CRM, ERP and human capital management systems. The company plans to build automated workflows that identify performance gaps and generate targeted training.

Sources

• 3 US AI Infrastructure Stocks Investors May Watch Beyond Texas Data Center Growth
• Sentira AI Launch Puts 10x Genomics Stock In Focus Before Earnings
• 3M (MMM) Expands AI Data Center Optics Push With 500 Job Investment
• What Is Broadcom (AVGO) Changing In AI Security And Fraud Defense?
• 3 AI Infrastructure Stocks Riding The Oil And Data Center Buildout
• Google Backs Unified Robots.txt Signal for AI Training 2026
• AI training opt out: How a new standard can help publishers protect content
• Parents Angry After School's AI Move With 6-Year-Olds
• Meta raises US$15 billion in five-part US bond sale to fund AI expansion
• Innodisk advances AI data center push with eSSD, DDR5 products
• EATS News: AI Framework Closes the Pilot Training Loop
• Daybreak Red is now available in AWS GovCloud (US-East)
• Copado Buys Valo.ai for Salesforce Security Observability
• Meta acquires AI coding startup Conductor
• Bites Raises $40 Million to Expand AI Training Technology for Restaurants and Connect Learning With Business Results |