Cisco has launched Antares, a family of small language models designed to help security teams identify vulnerabilities in code. The models, Antares-350M and Antares-1B, are open-weight and available on Hugging Face. They aim to provide a cost-effective solution for code security, reducing false positives and supporting data sovereignty.
Neo, a startup focused on securing AI agents, has raised $100 million in funding. The company aims to provide a real-time control layer for shifting software stacks, giving enterprises visibility and control over agentic software.
Anthropic, an AI research organization, has significantly increased its lobbying spending in the first half of 2026. The company aims to influence AI policy, particularly in emerging technologies and their potential impact on society.
Louisiana has enacted new laws regulating AI, which take effect on August 1. The laws expand criminal penalties for AI-generated sexual images, require disclosures on certain products and services, and impose new regulations on AI use in the state's justice system.
Key Takeaways
* Cisco launches Antares, a family of small language models for code security, available on Hugging Face.\n* Neo secures $100 million in funding for AI-powered security solutions.\n* Anthropic increases lobbying spending on AI policy.\n* Louisiana enacts new AI laws regulating AI use and penalizing AI-generated content.\n* Cisco's Antares models aim to reduce false positives and support data sovereignty.\n* Neo's platform offers real-time control and visibility over agentic software.\n* Anthropic's lobbying efforts focus on emerging AI technologies and their impact on society.\n* Louisiana's new laws take effect on August 1 and expand criminal penalties for AI-generated content.Cisco Unveils Antares AI Models for Secure Code Review
Cisco has launched Antares, a family of small language models designed to help security teams identify vulnerabilities in code. The models, Antares-350M and Antares-1B, are open-weight and available on Hugging Face. They aim to provide a cost-effective solution for code security, reducing false positives and supporting data sovereignty. Antares can help teams pinpoint vulnerable code, making it easier to prioritize and remediate security issues.
Cisco Launches Affordable AI-Powered Code Security
Cisco has introduced Antares, a small language model designed to help security teams identify vulnerabilities in code. Antares aims to provide a cost-effective solution for code security, reducing false positives and supporting data sovereignty. The model can help teams pinpoint vulnerable code, making it easier to prioritize and remediate security issues. Cisco claims Antares is up to 172x cheaper than competing models and can run on every commit.
Cisco Champions Smaller AI Models for Enterprise Security
Cisco is promoting smaller AI models, like Antares, for enterprise software security. These models help localize software vulnerabilities while reducing AI token costs and keeping sensitive source code on-premises. Antares-350M and Antares-1B are available as open-weight models on Hugging Face. They are designed to provide targeted capabilities with lower infrastructure costs and stronger data governance.
Neo Secures $100M for AI-Powered Security Solutions
Neo, a startup focused on securing AI agents, has raised $100M in funding. The company aims to provide a real-time control layer for shifting software stacks, giving enterprises visibility and control over agentic software. Neo's platform offers five core capabilities, including inventory cataloging, capability and risk intelligence, and real-time attribution.
Neo: AI Security for the Endpoint
Neo Security has secured $100M in funding to build AI security solutions for the endpoint. The company aims to provide enterprises with real-time visibility and control over their AI agents. Neo's platform answers key questions for CISOs, including what agentic software is being used, its security posture, and how to enforce effective controls.
CIOs Focus on AI ROI and Digital Transformation
CIOs gathered at the CIO 100 Leadership Live New York event to discuss AI ROI and digital transformation. Key areas of consensus included the need for solid IT fundamentals, strong business alignment, and a shift from project thinking to product thinking. CIOs warned that automating broken processes can amplify dysfunction and emphasized the importance of governance and measurement.
Louisiana Enacts New AI Laws
Louisiana has enacted new laws regulating AI, which take effect on August 1. The laws expand criminal penalties for AI-generated sexual images, require disclosures on certain products and services, and impose new regulations on AI use in the state's justice system. The laws aim to address the growing use of AI in the state and provide greater protections for citizens.
Anthropic Increases Lobbying Spending on AI Policy
Anthropic, an AI research organization, has significantly increased its lobbying spending in the first half of 2026. The company aims to influence AI policy, particularly in emerging technologies and their potential impact on society. Anthropic's increased lobbying efforts reflect the growing concerns and debates surrounding AI policy.
Byron Donalds on AI, Florida Governor Bid, and Data Centers
US Representative Byron Donalds discussed AI, his potential Florida governor bid, and data center infrastructure. He emphasized the need for sensible regulation and the importance of keeping American data within the US. Donalds also highlighted the national security implications of data warehousing and the need for robust infrastructure to support AI growth.
The Rise of Intimacy Economy and AI Companionship
Gen Z is struggling to find safety in human relationships, leading to the rise of the intimacy economy and AI companionship. AI chatbots and digital twins are filling the gap, offering emotional support and companionship. However, this trend raises concerns about the erosion of human emotional skills and the potential for emotional inequality.
China Adopts AI in Classrooms for Personalized Learning
Chinese schools are adopting AI to personalize learning, analyze student performance, and assist teachers. AI systems are being used to evaluate student code, provide feedback, and create customized exercises. Education officials stress that AI is intended to support, not replace, educators.
GRCA Partners with CatapalloVR for VR/AI Training
The Greater Reading Chamber Alliance (GRCA) is partnering with CatapalloVR to offer customized virtual reality and AI workforce training for Berks area businesses. The training aims to enhance employee skills and adaptability, supporting the growth and development of the local business community.
Diagnosing Semantic-Answerability Gap in Table RAG
Researchers have introduced TCR-Bench, a diagnostic benchmark for Table Content-level Answerability in RAG. The benchmark evaluates the ability of dense retrievers to retrieve relevant tables and pinpoint answerable information. The study highlights the Semantic-Answerability Gap, where retrievers struggle to identify uniquely answerable tables within similar groups.
Sources
- Introducing Antares: Highly Efficient Open Weight AI Models for Vulnerability Localization
- Cisco Launches Low-Cost AI Models for Source Code Security
- Cisco Makes the Case for Smaller AI in Enterprise Software Security as Antares SLMs Cut Token Costs
- Neo lands $100m as agentic AI outpaces security teams
- Neo: securing AI agents at the endpoint
- CIO 100 Leadership Live New York: CIOs push past AI pilots for measurable returns
- New Louisiana artificial intelligence laws take effect Aug. 1
- Anthropic ramps up lobbying spending amid AI policy fights
- Byron Donalds on AI, Florida Governor Bid & Data Centers
- Gen Z is living in an intimacy economy, where connection is commodified
- Artificial Intelligence Becomes Increasingly Common in Chinese Classrooms
- GRCA working with tech firm to offer training using VR/AI
- Semantically Similar, Logically Distinct: Diagnosing the Semantic-Answerability Gap in Table RAG
Comments
Please log in to post a comment.