microsoft announces salesforce

Key Takeaways

• The White House proposes a national AI framework to standardize regulations, protect children with enhanced controls, and address intellectual property rights.
• Enterprise AI agents, driven by platforms like Microsoft Copilot and embedded features in Salesforce, surged by 466.7% year-over-year, often operating without central governance and with administrative privileges.
• Accenture and Anthropic partnered to launch Cyber.AI, leveraging Anthropic's Claude AI model to provide continuous, AI-driven cybersecurity operations.
• OpenAI introduced a public safety bug bounty program to identify and address AI abuse and safety risks, including agentic risks like prompt injection.
• Tenable launched Hexa AI, an orchestration engine within Tenable One, to automate cybersecurity workflows and shift operations from reactive to proactive risk reduction.
• X-PHY offers hardware-enforced monitoring systems to secure AI agents, setting immutable limits on their actions outside the operating system's trust boundary.
• AI is assisting radiologists in prostate cancer detection, with Cleveland Clinic evaluating Siemens Healthineers' AI Rad Companion (AIRC) Prostate MRI for improved accuracy.
• The rapid growth of AI-powered tools in HR requires leaders to transform into strategic roles focused on talent strategy, or risk becoming obsolete.
• Teenagers are increasingly using AI chatbots for mental health support, raising concerns from experts about the chatbots' ability to address complex psychiatric issues with nuance.
• Researchers developed a new memristor using bismuth selenide (Bi2Se3) that combines long-term data retention and analog tuning, promising more efficient and faster AI hardware processing.

White House proposes national AI rules, child protection measures

The White House has proposed a national framework for artificial intelligence (AI) to create a single standard across the U.S. This plan aims to prevent states from creating their own AI laws that could be difficult for businesses. It also includes recommendations for protecting children online, such as stronger parental controls and age verification. The framework suggests that courts should decide how AI training on copyrighted material affects intellectual property. Companies must still navigate various state and federal rules until a national standard is set.

White House AI framework sparks national debate

The White House has released a new framework for artificial intelligence (AI) that suggests a national standard for AI development and use. This proposal aims to preempt state laws, but faces potential opposition in Congress. The framework emphasizes protecting children, including stronger parental controls and age verification measures. It also touches on intellectual property issues related to AI training data. While some lawmakers support a national approach, others express concerns about preemption, indicating a complex legislative path ahead.

White House AI policy framework detailed

The White House has outlined a national legislative policy framework for artificial intelligence (AI) focusing on seven key priorities. These include protecting children with better parental controls and age assurance, and safeguarding communities by addressing energy demands and fraud. The framework also addresses intellectual property rights, free speech, and promoting American AI innovation. It recommends preempting state AI laws that create undue burdens while preserving general state protections. The plan also emphasizes educating the workforce and developing AI talent.

Tenable Hexa AI streamlines security with automation

Tenable has launched Tenable Hexa AI, an orchestration engine within its Tenable One platform designed to automate cybersecurity workflows. This tool uses AI to manage cyber exposure by understanding how vulnerabilities, assets, and configurations interact across complex systems. It aims to move security operations from reactive responses to proactive risk reduction at machine speed. Hexa AI helps coordinate actions across IT, cloud, and AI environments, allowing security teams to focus more on critical tasks rather than manual upkeep.

Accenture and Anthropic partner for AI cybersecurity

Accenture and Anthropic have partnered to launch Cyber.AI, a new solution that uses Anthropic's Claude AI model to enhance cybersecurity operations. This platform aims to transform security responses from human speed to continuous AI-driven capabilities. Cyber.AI combines Accenture's cybersecurity expertise with Claude's reasoning engine to help organizations manage AI-related risks and operate at machine speed. The solution includes Agent Shield to protect, monitor, and govern autonomous AI agents in real-time, ensuring they align with company policies.

OpenAI launches safety bug bounty program

OpenAI has introduced a new public program to find and address AI abuse and safety risks across its products. This safety bug bounty program will reward researchers for identifying potential misuse or abuse that could lead to harm, even if it doesn't qualify as a traditional security vulnerability. It focuses on agentic risks like prompt injection and data exfiltration, as well as issues related to OpenAI's proprietary information and account integrity. The program aims to build a more secure AI ecosystem by collaborating with safety and security experts.

AI challenges HR to transform or become obsolete

The HR technology market is rapidly growing, with AI-powered tools set to automate many tasks currently performed by HR professionals. This presents a critical choice for HR leaders: either lead a transformation into a more strategic role or risk becoming a mere compliance function. AI's ability to automate content creation and analysis, beyond simple transactions, means HR must prove its strategic value. Leaders need to adapt by embracing AI, focusing on talent strategy, and guiding how employees interact with AI to elevate the HR function.

X-PHY secures AI agents with hardware

X-PHY CEO Camellia Chan discusses the security challenges posed by AI agents using Model Context Protocol (MCP) to access enterprise applications with elevated permissions. She explains that X-PHY's hardware-enforced monitoring and detection systems operate outside the operating system's trust boundary. This technology sets immutable limits on AI agent actions, preventing threats before data loss occurs. This allows organizations to confidently adopt agentic AI, especially as the ecosystem for tools like MCP has rapidly scaled since late 2024.

Gen Z, social media and AI shape fitness future

Artificial intelligence (AI), social media, and the Gen Z demographic are significantly influencing the future of fitness. Gen Z's preference for documenting workouts online fosters community and accountability. AI is enhancing how workouts are captured and shared, with tools like Samsung's Galaxy S26 Ultra Photo Assist helping to improve workout visuals by removing distractions. AI also aids in scheduling workouts and managing fitness-related communications, making it easier for individuals to maintain consistency and achieve their goals.

New memristor advances analog AI hardware

Researchers have developed a new memristor using bismuth selenide (Bi2Se3) that could significantly improve AI processing efficiency and speed. This memristor combines long-term data retention and analog tuning, crucial for hardware-based neural networks, without needing external regulators. Fabricated using a scalable method compatible with existing semiconductor manufacturing, the device allows for precise analog adjustments mimicking brain synapses. This breakthrough offers a promising pathway for creating more efficient AI hardware components.

AI agents surge in enterprises, study finds

A new analysis from BeyondTrust's Phantom Labs reveals a 466.7% year-over-year increase in AI agents operating within enterprise environments. These AI-driven identities, referred to as a 'shadow AI workforce,' are deployed without central governance or clear visibility into their privileges. The research highlights that many AI agents possess privileges similar to human administrators, expanding the identity attack surface. This rapid growth is fueled by AI platforms like Microsoft Copilot and embedded AI features in services like Salesforce and ServiceNow.

AI assists radiologists in prostate cancer detection

Cleveland Clinic researchers are evaluating AI software, the AI Rad Companion (AIRC) Prostate MRI from Siemens Healthineers, to help radiologists detect prostate cancer. This FDA-cleared technology assists in identifying and segmenting suspicious lesions on MRI scans, potentially improving detection accuracy and efficiency. The AI analyzes key image sequences, providing scores that aid urologic oncologists in biopsy and treatment decisions. This tool can help differentiate cancer from benign conditions and identify lesions in challenging locations, potentially reducing unnecessary biopsies.

AI chatbots offer teens mental health support

An increasing number of teenagers are using AI chatbots to seek mental health support. While these tools offer quick and accessible help, child and adolescent psychiatrist Meha Saxena from UConn Health expresses concern. She notes that AI chatbots may lack the necessary nuance to address complex psychiatric issues effectively. The availability of these AI resources raises questions about their suitability for providing in-depth mental health care to adolescents.

Sources

• White House National AI Policy Framework Calls for Preempting State Laws, Protecting Children
• White House Releases Long-Awaited Artificial Intelligence Framework, Setting the Stage for Federal Preemption Debate and Further Legislative Action
• In Summary: The White House National Legislative Policy Framework for Artificial Intelligence
• Tenable Hexa AI automates exposure management and security workflows
• Accenture and Anthropic Team to Help Organizations Secure, Scale AI-Driven Cybersecurity Operations
• Introducing the OpenAI Safety Bug Bounty program
• An AI Reckoning for HR: Transform or Fade Away | Brian Elliott | MIT Sloan Management Review
• X-PHY's Camellia Chan on hardware-enforced security for the age of AI agents
• How gen Z, social media and AI are shaping the future of fitness
• Memristor demonstrates use in fully analog hardware-based neural network
• Phantom Labs analysis of BeyondTrust’s Identity Security Insights Data finds enterprise AI Agents growing 466.7% year over year
• How AI Is Changing the Prostate MRI
• UConn Health Minute: AI and Teen Mental Health