Salesforce launches AI agents as Google releases Veo 3.1 Lite

Key Takeaways

• The Mississippi Artificial Intelligence Network (MAIN) and Valley City State University (VCSU) are offering free AI training and lectures focused on responsible AI use and its impact on education and the workforce.
• CompTIA advises companies to use accredited AI training rather than developing it in-house, highlighting cybersecurity as a primary AI-related challenge.
• SalesJack, an AI-powered CRM, helps lumber and building materials dealers identify and capture missed cross-sell opportunities by unifying data from existing ERPs and quoting tools.
• Amazon Web Services (AWS) launched AI agents for DevOps and security, which support multicloud environments, including AWS, Azure, and on-premises systems, to automate incident diagnosis and penetration testing.
• Google AI introduced Veo 3.1 Lite, a more affordable generative video model available via the Gemini API, offering HD video generation (720p at $0.05/sec, 1080p at $0.08/sec) with SynthID watermarking.
• Asana's enterprise AI agents are designed for collaborative workflows, integrating with third-party applications like Google Drive and Microsoft 365 to create shared institutional memory.
• AI tools can significantly assist overburdened regulatory systems by improving data analysis, risk assessment, and compliance checks, especially following the Supreme Court's Loper Bright decision.
• Neuroscientist James J. DiCarlo from MIT presented research showing that deep neural networks can model human vision and potentially be used to design images that elicit specific brain activity.
• Modern conflicts demonstrate that while advanced technology like AI is important, human factors, simpler tactics, and geographical advantages often play a decisive role.

State AI Initiative Teaches Responsible Use at USM

The Mississippi Artificial Intelligence Network (MAIN) held an AI workshop at the University of Southern Mississippi (USM) to teach people about responsible AI use. MAIN offers free AI training across Mississippi to students, educators, and government workers. USM aims to integrate ethical AI into teaching to prepare students for future careers. The workshop, attended by 450 people, also highlighted AI's potential as a brainstorming tool while emphasizing critical thinking to avoid misinformation. Students from other universities also attended, recognizing the need to adapt to AI's growing role in daily life.

VCSU Offers Free Public Lectures on AI's Impact

Valley City State University (VCSU) is hosting a free three-part lecture series in April called 'Making Sense of AI: Education, Workforce and Community.' These sessions, available in person and online, will explore how artificial intelligence is changing the workplace and the skills needed. Presented by VCSU's AI Institute for Teaching and Learning, the lectures aim to provide practical insights for students, parents, and professionals. Topics include an introduction to AI, its role in education, and its impact on the workforce.

SalesJack AI Helps LBM Dealers Find Missed Sales Opportunities

Many lumber and building materials (LBM) dealers miss cross-sell opportunities because customer and job data is scattered across different systems. SalesJack, an AI-powered CRM designed for LBM dealers, solves this problem by connecting to existing ERPs and quoting tools. It unifies data, cleans it, and uses AI to identify missing sales, such as quoting framing but not windows for the same job. SalesJack then sends automated alerts to sales reps, helping them capture overlooked revenue and improve win rates.

AWS AI Agents Automate DevOps and Security Tasks

Amazon Web Services (AWS) has launched AI agents to assist DevOps and security teams. The DevOps Agent acts as a site reliability engineer, investigating and diagnosing incidents, while the Security Agent performs autonomous penetration testing. Both agents now support multicloud environments, working across AWS, Azure, and on-premises systems. While they augment human teams by identifying issues and recommending solutions, they have limitations, such as limited write capabilities for the DevOps Agent and the need for human implementation of fixes. These agents are positioned as a cost-effective alternative to manual testing and high-salaried engineers.

CompTIA Advises Using Accredited AI Training

The Computing Technology Industry Association (CompTIA) advises companies not to develop AI training in-house, recommending instead the use of accredited and market-affiliated training materials. CompTIA research shows a significant need for AI training, with many companies still figuring out how to use the technology. Cybersecurity is identified as the biggest challenge related to AI, followed by the human factor, which requires education and policy. CompTIA integrates AI into its training modules and certifications to address the growing demand for AI skills in the workforce.

AI Tools Can Help Overburdened Regulatory Systems

Following the Supreme Court's Loper Bright decision, which ended agency deference, regulatory bodies face new challenges, especially with rapidly advancing AI. AI tools can significantly help these strained systems by improving data analysis, risk assessment, and compliance checks. They can also aid in predictive modeling and boosting overall agency efficiency. However, implementing AI requires expertise, infrastructure, and ethical guidelines, with concerns about bias and transparency needing careful consideration. Thoughtful AI adoption can help agencies become more precise and evidence-based in their regulatory actions.

Google's Veo 3.1 Lite Offers Cheaper, Faster AI Video Generation

Google AI has released Veo 3.1 Lite, a new, more affordable version of its generative video model available through the Gemini API and Google AI Studio. This model uses a Diffusion Transformer (DiT) architecture for better temporal consistency and operates in a compressed latent space for efficiency. Veo 3.1 Lite supports HD resolutions (720p and 1080p) and clip durations of 4 to 8 seconds, with pricing at $0.05 per second for 720p and $0.08 per second for 1080p. It also includes SynthID for watermarking AI-generated content, making it ideal for developers needing cost-effective, high-speed video generation.

Asana's AI Agents Focus on Collaboration for Enterprise

Asana's Chief Product Officer, Anne Raimondi, emphasizes that enterprise AI agents should be 'multiplayer by design,' operating within shared workflows rather than as standalone assistants. Asana's AI teammates access projects across the platform, allowing users to assign tasks, review outputs, and provide feedback, creating a shared institutional memory. These agents can connect to third-party apps like Google Drive and Microsoft 365, with more integrations planned. Asana believes this collaborative approach protects its platform's value against general-purpose AI agents by leveraging its Work Graph data layer.

AI Could Shape Future Human Vision, Lecture Explores

Neuroscientist James J. DiCarlo from MIT discussed how artificial intelligence can model and potentially shape human vision during a lecture at Stony Brook University. He explained that the brain's visual processing can be understood through deep neural networks, which now serve as scientific models mirroring how the brain works. DiCarlo's lab studies visual perception in humans and monkeys, finding that AI models align closely with neural data, offering a new paradigm for understanding intelligence. The research also explores using these AI models to design images that elicit specific brain activity patterns.

Technology's Role in Conflict: AI vs. Human Factors

Modern warfare highlights the complex interplay between advanced technology like AI and fundamental human factors. While technology can be decisive, as seen with Azerbaijan's use of drones against Armenia, relying solely on technological superiority can be dangerous. The conflict in Iran illustrates how simpler tactics and geographical advantages can counter advanced weaponry. The article suggests that human 'stupidity,' such as addiction to fossil fuels impacting global markets during geopolitical crises, remains a critical element in international conflicts, often outweighing technological might.

Sources

• State AI initiative comes to USM to teach responsible practices
• VCSU Hosting Free Public Lecture on (AI) Artificial Intelligence
• The Cross-Sell Blind Spot in LBM Sales and How SalesJack uses AI so Dealers Capture What They’re Missing
• AWS Deploys AI Agents To Do The Work Of DevOps And Security Teams
• AI training: No need to go it alone, says CompTIA
• AI Tools for an Overburdened Regulatory System
• Google AI Releases Veo 3.1 Lite: Giving Developers Low Cost High Speed Video Generation via The Gemini API
• Asana’s chief product officer: Why enterprise AI agents should be ‘multiplayer by design’
• Mind/Brain Lecture Explores How AI Could Shape Human Vision
• Artificial Intelligence vs. Human Stupidity