meta, openai and nvidia Updates

The artificial intelligence sector is currently experiencing rapid expansion, marked by significant investments, intense market competition, and increasing regulatory and security challenges. Experts like Danny Jenkins and Jon Clay from Trend Micro warn that AI-powered scams are becoming increasingly sophisticated and difficult to detect, often targeting businesses with malware. They advise caution with downloads and vigilance against messages pushing for quick actions, personal information, or unusual payments like gift cards or cryptocurrency. In response to these evolving threats, companies like Radware are developing new security measures, such as their LLM Firewall, which uses AI to protect large language models and generative AI applications by enforcing real-time security policies and detecting harmful prompts. Regulatory bodies are also scrutinizing AI's impact. Italy's antitrust watchdog, AGCM, has expanded its investigation into Meta Platforms over WhatsApp's AI tools and privacy policies. The probe examines whether Meta is blocking rival AI chatbots and if a new October 15 policy, allowing WhatsApp to share user data with Meta for AI training, constitutes unfair business practices. WhatsApp denies these claims, stating its business interface is not for AI chatbots, and is cooperating with the Italian authority. Meanwhile, OpenAI faces legal challenges, having lost a court battle requiring it to share internal communications regarding the deletion of two large collections of pirated books, a decision that could indicate knowing copyright infringement and lead to higher fines. Competition in the AI chip market is intensifying, with Nvidia facing growing pressure. A new challenger emerged recently, and Meta is reportedly exploring the use of Google AI chips, prompting Nvidia to assert the superior power and versatility of its technology. Despite these competitive pressures, Nvidia continues to highlight AI's global impact, as seen at its AI Day Seoul event, which showcased Korea's growing AI capabilities and recognized five local startups. Major companies like Samsung and Hyundai Motor Group are deeply integrating AI, strengthening Korea's digital foundation across various sectors. Investment in AI infrastructure remains a critical topic. OpenAI CEO Sam Altman suggests companies face a dilemma: overinvesting risks financial losses, while underinvesting could mean lost revenue opportunities. This echoes past concerns, such as Intel's historical overinvestments. Saudi Arabia is making ambitious moves to become a global AI leader by 2030, shifting its economic focus from oil and planning to fund extensive AI research, development, and data centers using its oil revenues. However, the country faces challenges, particularly the significant water requirements for cooling data centers in its desert climate. Across industries, AI adoption is accelerating. Retail and e-commerce leaders are significantly boosting AI spending, with 61% planning to implement agentic AI within the next year. They recognize generative AI's role in enhancing customer loyalty, powering conversational tools, and summarizing customer reviews. Increased budgets are anticipated for 2026, especially for hyper-personalization and predictive shipping. In healthcare, AI-based plaque assessments are gaining wider insurance coverage from providers like Medicare, Humana, United Healthcare, Cigna, and EviCore, with a tracking code set to become a CPT category 1 code in January 2026, signaling more permanent reimbursement and transforming coronary artery disease management.

Key Takeaways

• AI-powered scams are becoming highly advanced and difficult to detect, with experts advising caution against unusual payment requests and quick action prompts.
• Italy's antitrust watchdog is investigating Meta Platforms over WhatsApp's AI tools and a new policy allowing user data sharing with Meta for AI training.
• OpenAI must disclose internal communications regarding the deletion of pirated book collections, a ruling that could impact copyright infringement claims.
• Nvidia faces increasing competition in the AI chip market, with Meta reportedly considering Google AI chips.
• Radware launched a new LLM Firewall to protect large language models and generative AI applications from harmful prompts and data theft.
• OpenAI CEO Sam Altman highlights the dilemma for companies: overinvesting in AI infrastructure risks financial loss, while underinvesting risks lost revenue.
• Saudi Arabia aims to become a global AI leader by 2030, planning to fund AI research and data centers with oil revenues, despite water scarcity challenges.
• Retail and e-commerce leaders are significantly increasing AI spending, with 61% planning to implement agentic AI within the next year for customer experience.
• NVIDIA AI Day Seoul showcased Korea's growing AI power, with major companies like Samsung and Hyundai Motor Group integrating AI.
• Insurance reimbursement for AI-based plaque assessments is expanding, with Medicare and major private insurers now providing coverage, and a CPT category 1 code expected by January 2026.

Experts Warn AI Scams Are Getting Smarter and Harder to Spot

AI-powered scams are becoming very advanced and difficult to recognize, warn experts like Danny Jenkins and Jon Clay from Trend Micro. These scams often target businesses, aiming to infect many computers with malware. Victims might not realize what they see is fake, or they might download harmful software. Experts advise people to be careful about downloads and to check if messages push for quick actions, personal information, or unusual payments like gift cards or crypto.

New AI Scams Are Tricky Experts Say Be Careful

Experts like Jenkins and Clay warn that AI-driven scams are becoming much harder to detect. Many people do not realize AI can create such convincing fakes, making victims vulnerable. Attackers often target businesses to gain access to many computers through malware. People should be very careful about what they download. Always check if a message tries to make you act quickly, share personal information, or ask for unusual payments like gift cards or crypto.

AI Scams Get Smarter Experts Offer Safety Tips

Experts like Jenkins and Clay warn that AI-driven scams are getting much harder to spot. Many victims do not know that what they see might not be real. Attackers often target businesses to get access to many computers by tricking people into downloading malware. Experts advise everyone to be very careful about downloads. Also, check if a message asks you to do something quickly, share personal information, or make unusual payments like gift cards or crypto.

Italy Expands Probe Into Meta Over WhatsApp AI Tools

Italy's antitrust watchdog is expanding its investigation into Meta Platforms regarding its AI tools in WhatsApp. The watchdog worries Meta might be blocking rival AI chatbots from using its popular messaging service. Italy may put temporary rules in place, like stopping new terms or limiting Meta AI in WhatsApp. WhatsApp denies these claims, stating its business interface is not for AI chatbots. The regulator also notes Meta changed WhatsApp Business terms on October 15, banning some AI service companies. This investigation will finish by the end of 2026.

Italy Expands WhatsApp AI Data Sharing Investigation

Italy's competition authority, AGCM, has expanded its investigation into WhatsApp's privacy policy and how it uses AI. The probe now includes a new October policy that lets WhatsApp share user data with Meta for AI training. AGCM will also examine how WhatsApp's own AI features handle user data. The authority wants to see if these new terms and AI integrations are unfair business practices. WhatsApp states it is working with the Italian authority and believes its actions follow all laws.

OpenAI Must Share Details About Deleted Pirated Books

OpenAI lost an important court battle against authors suing the company. A judge ruled that OpenAI must share internal messages about deleting two large collections of pirated books, known as "books 1" and "books 2." This decision could show that OpenAI knowingly broke copyright laws, which means much higher fines. OpenAI had claimed the books were deleted because they were not used, but the court said this opened the door for discovery. This ruling helps authors who argue that illegally downloading works is copyright infringement.

Nvidia Faces Growing Pressure in AI Chip Market

Nvidia's leading position in the AI chip market is becoming less secure, facing concerns over its high value. A new challenger appeared last week, and Meta is reportedly looking to use Google AI chips. Nvidia responded on X, stating its technology is more powerful and versatile than other AI chips. While U.S. stocks advanced on Tuesday, economic statistics were downbeat. Meanwhile, a Ukraine peace plan is moving forward, and the UK is preparing its Autumn Budget.

Radware Launches New Firewall to Secure AI Models

Radware has created a new LLM Firewall to protect large language models and generative AI applications. This firewall, an add-on to their Cloud Application Protection Services, enforces security policies in real time. It uses an AI model to detect harmful prompts or attempts to steal personal information, addressing top risks identified by OWASP. Dror Zelber, a Radware VP, explains that the company is using AI to secure AI, as attacks on these new technologies will be hard to predict. Many organizations are not yet fully prepared for these cybersecurity risks.

Can Tech Companies Overspend on AI Infrastructure

Tech companies are heavily investing in computing infrastructure for artificial intelligence. This article questions if it is possible to spend too much on AI, drawing a parallel to Intel's past overinvestments. OpenAI CEO Sam Altman believes companies face a choice: either overinvest and lose money, or underinvest and lose potential revenue. Pat Gelsinger, who tried to turn Intel around, retired in late 2024.

Saudi Arabia Aims for AI Leadership Faces Challenges

Saudi Arabia wants to become a global leader in artificial intelligence by 2030, despite currently trailing China and the U.S. The country is shifting its focus from oil to AI and high-tech cooperation with the U.S. Prince Mohammed bin Salman recently discussed tech during a White House visit. Saudi Arabia plans to use its large oil revenues to fund AI research, development, and data centers. However, a major challenge is the need for significant water to cool data centers, which is scarce in the desert kingdom.

Retailers Boost AI Spending for Customer Experience

Retail and e-commerce leaders plan to keep investing in artificial intelligence, according to a new survey by Algolia. Most organizations, 61%, will implement agentic AI within the next year, driven by easy integration and return on investment. Nearly all leaders believe generative AI helps customer loyalty and powers helpful conversational tools. Many also value AI's ability to summarize customer reviews. Spending on AI-powered search grew this year, and more businesses plan to increase their AI budgets in 2026, especially for hyper-personalization and predictive shipping.

NVIDIA AI Day Highlights Korea's Growing AI Power

NVIDIA's AI Day Seoul event gathered over 1,000 attendees to explore Korea's growing artificial intelligence landscape. Experts discussed new AI trends, including agentic and physical AI, and held workshops. Five Korean startups were recognized as finalists in the Inception Grand Challenge, receiving awards and certificates. NVIDIA leaders noted that AI is strengthening Korea's digital foundation, with major companies like Samsung and Hyundai Motor Group deeply integrating the technology. AI is also helping government services, gaming, music, and the K-pop industry.

AI Plaque Assessments Get Better Insurance Coverage

Insurance reimbursement for AI-based plaque assessments is improving, making this technology more accessible. Medicare already covers these assessments for certain patients, and major private insurers like Humana, United Healthcare, Cigna, and EviCore are now also providing coverage. Dr. Parikh notes that this progress is supported by strong evidence showing how CT-based plaque assessment helps manage patient care. A tracking code will become a CPT category 1 code in January 2026, signaling more permanent reimbursement. This wider coverage will help transform how coronary artery disease is assessed and managed.

Sources

• AI-driven scams growing in sophistication, experts warn
• AI-driven scams growing in sophistication, expert warns
• AI-driven scams growing in sophistication, expert warns
• Italy broadens antitrust probe into Meta over AI tools in WhatsApp
• Italy’s Competition Watchdog Broadens WhatsApp AI Policy Probe
• OpenAI Loses Key Discovery Battle as It Cedes Ground to Authors in AI Lawsuits
• CNBC Daily Open: Nvidia's crown looks increasingly uneasy on its head
• Radware Adds Firewall for LLMs to Security Portfolio
• It Really Is Possible to Spend Too Much on AI
• Saudi Arabia’s Push For AI Remains A Challenge
• Survey: Retail, e-commerce execs remain bullish on AI
• From Government to Gaming, AI Is ‘Strengthening Korea’s Digital Foundation,’ NVIDIA Leader Says at AI Day Seoul
• Reimbursement for AI-based plaque assessments is improving