OpenAI deploys ChatGPT as Anthropic aids missions

Key Takeaways

• The Pentagon is pushing OpenAI and Anthropic to deploy AI tools, including ChatGPT, on classified military networks for intelligence and mission planning, balancing national security with ethical concerns.
• AI-powered romance scams led to over $672 million in losses for Americans in 2024, with men being 65% more likely to encounter them weekly.
• AWS CEO Matt Garman believes fears of an AI-driven software stock sell-off are "overblown," while Amazon Web Services showed strong cloud growth.
• The global Data Center Accelerator market is projected to grow 25% annually for five years, driven by AI investments, according to Dell'Oro Group.
• Senator Bernie Sanders' proposal to halt new AI data center creation was criticized as a "power grab" by economist Peter St. Onge.
• An AI camera on Minnesota's Highway 7 has reduced traffic fatalities to zero and halved serious injury crashes since its installation.
• Black Badger Software Solutions launched AI training on February 12, 2026, to help non-technical business teams integrate AI for automation and risk reduction.
• UK AI course enrollment grew 36% in 2024-25, highlighting the need for flexible and frequently updated academic programs.
• Tech CEO Matt Shumer warned on February 12 that the public is unprepared for the rapid advancement of artificial intelligence.
• Senators alleged that the AI toy Miko exposed thousands of children's conversation responses in an unsecured database, raising significant privacy concerns.

Pentagon urges AI firms to use tools on secret networks

The Pentagon is pushing top AI companies like OpenAI and Anthropic to make their artificial intelligence tools available on classified networks. They are asking for fewer standard restrictions on these tools to speed up the military's adoption of advanced AI. This effort aims to maintain a technological advantage over rivals such as China by leveraging AI for tasks like intelligence analysis and operation planning. Discussions are ongoing, with companies balancing lucrative government contracts against ethical and security concerns. The outcome could significantly impact the future of AI in national security.

Pentagon seeks AI tools for secret military networks

On February 11, the Pentagon urged top AI companies like OpenAI and Anthropic to make their AI tools available on classified networks without standard restrictions. Pentagon Chief Technology Officer Emil Michael stated the military aims to deploy AI models across all security levels. This effort is part of ongoing talks about using AI on future battlefields, but companies have concerns about restrictions and potential mistakes. While OpenAI reached a deal for unclassified use, Anthropic is already used for national security missions and is in discussions with the Department of War. Military officials want to use AI to synthesize information for decisions, but researchers warn of deadly consequences if AI makes errors in classified settings.

Pentagon seeks unrestricted AI use on secret systems

The Pentagon wants to use the newest AI models on both unclassified and classified networks without restrictions. Defense officials are pushing for advanced AI tools to be used for sensitive tasks like mission planning and potentially weapons targeting. Secretary of War Pete Hegseth stated the military is fully committing to artificial intelligence. While OpenAI's ChatGPT is used for daily tasks on an unclassified network, Anthropic has expressed concerns about its models being used for autonomous weapons. These negotiations are difficult for AI companies, who must balance lucrative government contracts with ethical concerns and employee opposition.

Spotting AI romance scams and avoiding financial loss

Americans lost over $672 million to romance scams in 2024, with AI making these scams more common. Men are 65% more likely to encounter these scams weekly, and only one in four victims recover their money. Scammers use social media to find victims, build trust by mirroring interests, then ask for money through hard-to-trace methods like gift cards or cryptocurrency. Cyber news reporter Kerry Tomlinson advises looking for red flags such as sudden changes to "flowery" writing in messages. AI-generated photos can be spotted by looking for strange details in backgrounds or on faces and hands. In video calls, asking someone to touch their face or turn their head can sometimes reveal a deepfake, though AI technology is quickly improving.

Learn to identify AI romance scam red flags

Romance scams, now powered by AI, caused Americans to lose over $672 million in 2024. Men are 65% more likely to encounter these scams weekly, and few victims recover their lost money. Scammers build trust on social media by creating fake personas and mirroring victims' interests before asking for money through untraceable methods. Expert Kerry Tomlinson advises looking for unusual "flowery" language in messages as a sign of AI use. AI-generated photos often have strange details in the background or on body parts like extra fingers. During video calls, asking the person to move their hand to their face or turn their head can sometimes reveal a deepfake, but AI is rapidly improving.

Economist calls Sanders AI data center plan a power grab

Peter St. Onge, a senior economist at The Heritage Foundation, discussed Senator Bernie Sanders' proposal. Senator Sanders, I-Vt., suggested stopping the creation of new AI data centers. St. Onge criticized this idea on "The Bottom Line," calling it a "power grab."

AWS CEO says AI software fears are too high

AWS CEO Matt Garman believes fears about AI causing a software stock sell-off are "overblown." Analysts have called the drop in software as a service stocks a "SaaS apocalypse." Garman acknowledged AI is a disruptive force that will change how software is used and built. He stated that current SaaS providers and large companies can succeed if they continue to innovate. Amazon Web Services itself showed faster cloud growth than expected in the fourth quarter.

Tech CEO warns public unprepared for fast AI growth

On February 12, a tech CEO named Matt Shumer shared a warning on "The Lead." He stated that the public is not ready for how quickly artificial intelligence is advancing. Shumer's comments highlight concerns about the rapid development of AI technology.

AI degrees need flexibility as student numbers rise

The number of students taking AI courses in the UK grew by 36% in 2024-25, reaching 10,995, making it one of the fastest-growing subjects. Experts warn that these rapidly evolving courses need to be flexible and updated often to stay relevant. Professor Leo McCann from the University of Manchester suggests combining traditional degrees with shorter, modular courses. Risks include AI content quickly becoming outdated and ensuring academic programs have enough depth. Over half of the AI students in 2024-25 were international, with most enrolled in postgraduate courses.

AI camera reduces deaths on Minnesota Highway 7

An AI camera installed on Highway 7 in Minnesota has helped reduce traffic fatalities and serious injuries. In 2024, there were five deaths on the highway, but since the camera was added a year ago, there have been zero fatalities. Sergeant Adam Moore of the South Lake Minnetonka Police Department reported that the camera led to 1,200 distracted driving stops, a big increase from 300 before. The Highway 7 Safety Coalition also noted that serious injury crashes have been cut in half. The camera costs $15,000 per month to lease, and officials hope to secure a grant to keep it operating and potentially expand its use to other dangerous roads.

Black Badger offers AI training for business owners

On February 12, 2026, Black Badger Software Solutions launched a new training platform for business owners. This practical AI training helps non-technical teams use AI in their daily work without needing to hire developers. Founder David Verneuille emphasizes building AI systems that truly help teams, not just playing with tools. The program teaches how to identify tasks AI can automate, like reporting and data cleanup, and how to set boundaries so AI supports staff decisions. It also focuses on reducing risks by building review steps and connecting AI to existing business systems like project management and CRM. The training uses practical exercises to help businesses implement AI in a controlled way, improving operations and saving time.

AI investments drive data center growth says Dell'Oro

On February 12, 2026, Dell'Oro Group reported that the global Data Center Accelerator market will grow by 25% each year for the next five years. This growth is mainly due to the increasing use of AI applications, which need special infrastructure with accelerators. Baron Fung, a Senior Research Director at Dell'Oro Group, stated that AI is starting a new phase of expansion for server and storage components. Hyperscale cloud providers are using custom accelerators to improve efficiency and lower costs. The strong demand for memory and storage is expected to create supply shortages and higher costs, pushing companies to develop more efficient designs.

Senators say AI toy Miko exposed kids' conversation data

Senators Marsha Blackburn and Richard Blumenthal claim that the AI toy maker Miko exposed thousands of its responses to children's questions. Staff from the senators' offices found that Miko's audio responses were in an unsecured, publicly available database. This database, viewed by NBC News, appeared to contain Miko's side of conversations with children, sometimes including children's names and details. Miko CEO Sneh Vaswani denied any data breach, stating the company does not store children's voice recordings or personal information. However, experts like Miranda Bogen from the Center for Democracy and Technology expressed concerns about the lack of security for interactions with AI systems. The senators are questioning whether Miko adequately protects children's privacy and security.

Sources

• Exclusive-Pentagon pushing AI companies to expand on classified networks, sources say
• Exclusive: Pentagon pushing AI companies to expand on classified networks, sources say
• The Pentagon Wants to Raw Dog the Latest AI Models on Classified Systems
• Men are more likely to fall for this scam. Here's how to spot the red flags
• Men are more likely to fall for this scam. Here's how to spot the red flags
• Economist emphasizes Bernie Sanders’ stance on AI data centers is a ‘power grab’
• AWS CEO Garman says software AI fears are 'overblown'
• Tech CEO warns public isn’t ready for rapidly advancing AI
• ‘Flexibility needed’ to keep AI degrees relevant as demand soars
• No deaths, serious injuries down since AI camera added on Highway 7 in Minnesota
• Black Badger Software Solutions Launches Practical AI Training for Business Owners
• Hyperscale AI Investment Cycle Anchors Accelerator-Led Infrastructure Spending, According to Dell'Oro Group
• AI toy maker exposed thousands of responses to children, senators say