Meta Platforms used its Meta Connect 2026 event to unveil a wave of new hardware, including VR glasses, Ray-Ban smart glasses, and the Muse Charm AI agent. The company is backing its push with a massive $130 billion to $145 billion AI capital expenditure plan for 2026. A core goal is reducing its reliance on outside suppliers by developing its own AI hardware, a move that could give Meta more control over production and efficiency. Other tech giants like Amazon and Google are also investing heavily in AI research.
Despite Meta's ambitions, Apple investors appear largely unfazed. iPhone sales rose over 20% year over year in the first nine months of fiscal 2026, and the iPhone is expected to remain the leading device for AI interaction. Meta has 3.6 billion daily active users across its apps and earns most of its revenue through ads, but consumer habits are hard to break. A report suggests Meta's hardware push will not threaten Apple's dominance in the near term.
On the consumer side, a survey of 2,000 U.S. adults by Aura found that people are sharing sensitive details with AI chatbots in alarming numbers. About 73% said they shared something with AI they would not post publicly, and 71% said they told AI something they would be embarrassed for a close friend to know. Even though 86% of respondents worried that AI conversation data could be exposed in a breach, 81% continued sharing sensitive information anyway.
Cybersecurity experts warn that these conversations can reveal details about health, finances, and relationships, potentially building a profile used for targeted scams. They recommend keeping passwords and Social Security numbers out of AI chats and checking privacy settings regularly. Older adults face even steeper challenges: a Senior Planet from AARP survey found only 7% feel very confident protecting personal information online, and 70% lack confidence in knowing what information is safe to share with AI tools.
In the infrastructure and policy space, Nvidia launched a new AI agent security platform that puts hardware at the center of agent control. The BlueField DPU is seeing growing demand as AI adoption increases, but enterprises must balance stronger protection against higher infrastructure costs and deeper dependence on Nvidia's stack. Meanwhile, Google's release of Gemini 4 Argon, a model designed for complex coding and defensive cybersecurity, has sparked a Senate debate on AI liability. Senators Josh Hawley and Chris Murphy plan to introduce bipartisan legislation that would hold AI companies responsible for hacking incidents, similar to bank cyber attack accountability.
In other developments, SAS hired Casey McGee as Executive Vice President and Chief Sales Officer, effective October 1, 2026. McGee previously held leadership roles at Microsoft working with ISVs and system integrators. The move targets SAS's weaker ecosystem alignment and comes as over 50% of enterprise decision-makers now prioritize generative and agentic AI tools. Separately, a new method called Reserve-C4B helps conservative bandit algorithms improve existing policies without exceeding performance budgets, while another study reveals that a classifier's accuracy can drift even when its confidence distribution stays the same.
On the traffic and accessibility front, Jim Cramer says Akamai is the only AI traffic stock worth buying, citing its low valuation and an $11.6 billion deal with Anthropic. Akamai is building smaller data centers to process AI workloads closer to users, while Fastly is also shifting toward AI-focused services. In a heartwarming story, Dave Cragg, a 65-year-old blind chicken farmer in Cheshire, received AI-powered Meta smart glasses for his birthday. The glasses let him read books, menus, and birthday cards aloud and help him navigate using connected apps, making him much more independent in daily life.
Key Takeaways
- <ul><li>Meta plans to spend $130 billion to $145 billion on AI capital expenditures in 2026</li><li>Meta is developing its own AI hardware to reduce reliance on third-party suppliers</li><li>iPhone sales rose over 20% year over year in the first nine months of fiscal 2026</li><li>73% of U.S. adults shared something with AI they would not post publicly, per an Aura survey</li><li>Nvidia launched a new AI agent security platform centered on BlueField DPU hardware</li><li>Senators Hawley and Murphy plan to introduce bipartisan AI liability legislation</li><li>SAS hired Casey McGee, formerly of Microsoft, as Chief Sales Officer effective October 1, 2026</li><li>Akamai secured an $11.6 billion deal with Anthropic and is building smaller data centers for AI workloads</li><li>Only 7% of older adults feel very confident protecting personal information online</li><li>AI-powered Meta smart glasses helped a blind farmer in Cheshire gain daily independence</li></ul>
Meta Launches New AI Hardware but iPhone Dominance Likely Safe
Meta Platforms unveiled new devices at Meta Connect 2026, including VR glasses, Ray-Bin smart glasses, and the Muse Charm AI agent. The company plans to spend $130 billion to $145 billion on AI capital expenditures in 2026. Despite Meta's push, the iPhone remains the leading device for AI interaction and sales were up over 20% year over year in the first nine months of fiscal 2026.
Meta Invests in AI Hardware to Reduce Reliance on Outside Suppliers
Meta Platforms is developing its own AI hardware to reduce dependence on third-party suppliers and gain more control over production. The move aims to give Meta a competitive edge in the AI market by offering more efficient solutions. Experts note the technology is still in early development. Other companies like Amazon and Google are also investing heavily in AI research.
Meta's AI Hardware Plans Won't Threaten Apple Investors, Report Says
Meta Platforms announced new hardware at Meta Connect 2026, including VR glasses and the Muse Charm device. The company has 3.6 billion daily active users across its apps and earns revenue mainly through ads. However, consumer habits are hard to change, and the iPhone is expected to remain the top device for AI interaction. iPhone sales rose over 20% year over year through the first nine months of fiscal 2026.
Survey Reveals People Share Sensitive Details with AI Chatbots
A survey of 2,000 U.S. adults by Aura found that 73% shared something with AI they would not post publicly. 71% said they told AI something they would be embarrassed for a close friend to know. 86% considered that AI conversation data could be exposed through a breach. Even among concerned users, 81% continued sharing sensitive information with AI tools.
Experts Warn AI Chatbot Use Can Expose Personal Information
Cybersecurity experts say AI chatbot conversations can reveal sensitive details about health, finances, and relationships. Aura found that 71% of people shared with AI something they would be embarrassed to tell a friend. Experts warn that this information can build a profile used for targeted scams. They recommend keeping passwords and Social Security numbers out of AI chats and checking privacy settings regularly.
Nvidia AI security platform raises hardware cost concerns
Nvidia launched a new AI agent security platform that puts hardware at the center of agent control. The BlueField DPU demand is expected to grow with AI adoption. However, enterprises must weigh stronger protection against higher infrastructure costs and deeper dependence on Nvidia's stack.
New method improves conservative bandit algorithm reliability
Researchers developed Reserve-C4B to help conservative bandits improve existing policies without exceeding performance budgets. The method addresses shared estimation errors when baselines are uncertain. It uses a reserve ledger to separate statistical evidence from allowed performance deficits. Reproducible experiments showed large reductions in baseline fallback while exposing limitations of frozen certificates.
Google Gemini 4 Argus model sparks Senate AI liability debate
Senators Josh Hawley and Chris Murphy plan to introduce bipartisan legislation holding AI companies liable for hacking incidents. Google released Gemini 4 Argon, a model designed for complex coding tasks and defensive cybersecurity. The bill would make AI companies responsible for hacking incidents similar to bank cyber attack accountability. This comes as the Trump administration increases scrutiny of AI companies.
SAS hires Casey McGee to boost AI sales and partnerships
SAS named Casey McGee as Executive Vice President and Chief Sales Officer effective October 1, 2026. The hire targets SAS's weak ecosystem alignment scored at 3.60 and go-to-market execution at 4.62. McGee previously held leadership roles at Microsoft working with ISVs and system integrators. Over 50 percent of enterprise decision-makers now prioritize generative and agentic AI tools.
Older adults struggle with AI privacy and online safety
A Senior Planet from AARP survey found only 7 percent of older adults feel very confident protecting personal information online. Seventy percent said they lack confidence in knowing what information is safe to share with AI tools. Tom Kamber from Older Adults Technology Services discussed password protection, recognizing suspicious websites, and careful sharing of personal information with AI.
Akamai is the only AI traffic stock worth buying says Cramer
AI agents are creating more web traffic, which helps internet companies like Cloudflare, Akamai and Fastly. Jim Cramer says Akamai is the only one worth buying because of its low valuation and an $11.6 billion deal with Anthropic. Akamai is building smaller data centers to process AI workloads closer to users. Fastly is also shifting toward AI-focused services.
Study examines reliability drift under fixed confidence distribution
A classifier's accuracy can change even when its confidence distribution stays the same. Researchers studied how reliability can shift under covariate shifts that preserve confidence score distribution. They found that calibration residual and grouping variance do not always determine fragility. Tests on ImageNet showed positive bounds for several classifiers.
AI smart glasses help blind farmer gain independence
Dave Cragg, a 65-year-old chicken farmer from Cheshire, lost his sight at age 10 and has very limited peripheral vision. He received AI-powered Meta smart glasses for his birthday, which let him read books, menus and birthday cards aloud. The glasses also help him navigate using connected apps and find objects around him. His daughter says the technology has made him much more independent in daily life.
Sources
- Meta Platforms Is Making a Push Into AI Hardware. History Says Apple Investors Shouldn't Worry.
- Meta Platforms Is Making a Push Into AI Hardware. History Says Apple Investors Shouldn't Worry.
- Meta Platforms Is Making a Push Into AI Hardware. History Says Apple Investors Shouldn't Worry.
- AI, TMI? Users Tell Chatbots What They Wouldn't Want Their Closest Friends to Know
- Cybersecurity experts warn AI chatbot conversations could expose sensitive information
- Focus: The hidden price of Nvidia's AI agent security play
- Reserve-Aware Contrast Certificates for Conservative Bandits with Uncertain Baselines
- AI liability fight reaches Senate as Google releases powerful new model
- SAS Bets on Partner Expertise to Close Its AI Go-to-Market Gap
- Protecting your privacy in the AI era
- AI is a boon for 3 internet traffic cop stocks. Cramer says only 1 is a buy right now
- How Much Can Reliability Drift Under a Fixed Confidence Distribution?
- 'AI glasses give me independence as a blind farmer'
Comments
Please log in to post a comment.