scale ai launches nvidia while google expands its platform

Key Takeaways

• Lawsuits are challenging the use of AI to create nonconsensual deepfake pornography, testing existing revenge porn laws.
• AI nearly caused a $50 million penthouse real estate deal to collapse, highlighting its potential impact on high-value transactions.
• Federal courts have issued conflicting rulings on whether AI-generated materials, including communications with ChatGPT, are protected by legal privilege.
• Grip Security CTO Idan Fast argues that effective AI governance requires managing access and permissions for AI agents interacting with business systems.
• Enterprise AI agents need rigorous stress testing, not just demonstrations, to ensure reliability and prevent issues like compliance violations.
• Nvidia CEO Jensen Huang predicts that artificial intelligence will create more jobs, similar to past technological advancements.
• The University of Kent is hiring an AI Training Specialist to integrate generative AI tools into its educational and operational frameworks.
• Senator Tom Cotton dismissed claims that Google Gemini flagged his comments as hate speech, with Google emphasizing Gemini's low bias and anti-discrimination policies.
• AI is influencing military planning, raising concerns among experts about its potential to favor specific operational thinking.
• AI-powered fitness mirrors, such as those from Magic AI, are being explored as a new frontier in the home fitness market.

Arizona women sue over AI porn using their images

Three women in Arizona are suing companies and individuals, claiming their photos from Instagram were used to create fake pornographic videos. They filed the lawsuit under Arizona's revenge porn law because the state, like many others, does not yet have a ban on creating nonconsensual sexual deepfakes. The lawsuit states that some defendants created a company to help users make AI porn and profit from it while avoiding legal trouble. Their attorney believes they can prove the AI porn was made with the intent to harm, and this case could test if current revenge porn laws apply to AI-generated sexual content.

Kansas City woman joins lawsuit against AI porn creators

Three women, including one from Kansas City, have filed a lawsuit claiming that AI platforms were used to create fake pornographic videos using their likenesses from Instagram posts. They are suing under Arizona's revenge porn law, as many states, including Missouri, have not yet banned the creation of nonconsensual sexual deepfakes. The lawsuit alleges that some defendants formed a company to produce and profit from AI-generated pornography while evading legal consequences. Their lawyer, Nick Brand, notes that victims often don't know they've been targeted until the explicit content is seen by others.

AI nearly ruined $50 million penthouse deal

Real estate mogul Ryan Serhant shared on 'The Bottom Line' how artificial intelligence almost caused a $50 million penthouse sale to fall through. The details of how AI interfered with the high-stakes transaction were discussed during the program.

University of Kent hires AI training expert

The University of Kent is hiring an AI Training Specialist to support its AI@Kent initiative. This new role will focus on providing practical training for staff and students on using generative AI tools. The goal is to move beyond policy discussions and integrate AI into teaching, assessment, and daily university operations. The specialist will develop training sessions, create guidance materials, and help academic teams adapt to the growing use of AI in education.

CNN producer tries AI fitness mirror

CNN producer Leroy Ah-Ben tested Magic AI, a London-based company's AI-powered fitness mirror. He explored the technology, its competitors, and whether AI can truly deliver on the fitness industry's promises. The report examines if this computer-vision mirror can succeed in the home fitness market.

AI agents need testing not just demos

Abhishek Saxena of Sentient argues that enterprise AI agents require thorough stress testing, not just impressive demos, to prove their readiness for critical tasks. He explains that AI agents can cause serious issues like compliance violations or financial losses if they fail in real-world, unpredictable situations. Saxena emphasizes that the focus should be on building robust testing infrastructure to evaluate how AI agents handle ambiguity and pressure, ensuring they are trustworthy before deployment.

Conflicting court rulings on AI and legal privilege

Two recent federal court decisions have reached different conclusions about whether AI-generated materials are protected by attorney-client privilege or work product doctrine. One court ruled that communications with ChatGPT are not privileged because the AI could be seen as a third party. However, another court found that communications with an AI tool used by Zillow were protected under the work product doctrine. These conflicting rulings create uncertainty for businesses and legal teams regarding the confidentiality and discoverability of AI-generated information.

AI changes military planning, experts say

As artificial intelligence becomes more integrated into military planning, practitioners are experiencing unease despite improved tools and data access. Experts suggest that AI may be favoring certain ways of thinking about operations, potentially obscuring others. This tension arises as AI shapes what planners see and consider relevant, leading to challenges in achieving coherence and a shared sense of direction in campaigns. The discussion highlights the need to understand how different reasoning methods coexist and how AI influences operational art.

AI governance needs access control, not just model checks

Grip Security CTO Idan Fast states that effective AI governance starts with managing access, not just focusing on AI models themselves. He explains that AI agents gain power through their ability to access and act within business systems, which is controlled by identity and permissions. Fast argues that the main challenge for enterprises is not the AI models but the security risks arising from how AI agents connect to and interact with various SaaS applications. He stresses the need for governance frameworks that provide visibility into these connections to manage AI risks.

Senator Cotton dismisses Google AI hate speech claim

Senator Tom Cotton has dismissed a report claiming Google Gemini accused him of violating hate speech policies. The report stated Gemini flagged comments made by Cotton on gay rights, diversity, and transgender sports. Cotton's office has stated that the vast majority of Americans agree that girls' sports should be for girls. Google maintains that Gemini is among the least biased AI models and works to mitigate bias through testing, while also having policies against bullying and discrimination.

Nvidia CEO: AI will create jobs

Nvidia CEO Jensen Huang believes artificial intelligence will create more jobs rather than eliminate them. He compares AI's impact to that of personal computers and the internet, which also increased productivity and created new work opportunities. Huang pointed out that there is a significant shortage of workers in fields like truck driving and manufacturing, suggesting that AI will lead to increased demand for labor.

Sources

• Arizona women's deepfake lawsuit targets AI porn industry
• Women's deepfake lawsuit targets AI porn industry
• How AI nearly derailed $50 million real estate deal
• University of Kent recruits AI training specialist for AI@Kent | ETIH EdTech News
• Watch CNN producer take on an AI workout mirror
• Enterprise AI Agents Need Stress Tests, Not Sales Pitches
• AI, Privilege, and Work Product: Conflicting Federal Decisions Create a New Risk Frontier
• Military Operational Thinking in an Age of Artificial Intelligence
• AI Governance Starts With Access, Not Models
• Republican senator dismisses reported violation of AI hate speech policy
• Nvidia's Huang says AI will create jobs, not cut them