🪖 Chinese military AI

🧵 In today's edition:

• 🪖 Chinese military AI

• 🧑‍💻 AI code

• 🖼️ Agentforce AI

• 🛠️ AI Tools to Check Out

• 🤑 AI Fundraising News

• 🐦 Tweet Post of the Day

🪖 Chinese military AI

According to recent academic papers, Chinese researchers have reportedly adopted Meta’s Llama AI model for military applications.

Here’s the TL;DR:

• These researchers, associated with the People's Liberation Army’s (PLA) Academy of Military Science, have created an AI tool called "ChatBIT" by fine-tuning an older version of Llama to support military decision-making, intelligence analysis, and strategic planning.

• Meta’s open-source models come with usage restrictions barring military and warfare applications, but enforcing these conditions is challenging given the model's public accessibility.

What was Meta’s reaction?

• Meta responded by reaffirming that any military usage of their models by the PLA is unauthorized.

• Meta also stressed that the global AI race is driven by massive investments, including China's efforts, and called for continued open innovation in the U.S. despite security risks.

• The Pentagon is monitoring foreign AI developments and considering the implications for national security.

Overall: The U.S. government has recently introduced policies aimed at limiting technology investments in China, reflecting concerns over national security. AI experts, however, highlight the difficulty of keeping advanced technology from China, where ongoing partnerships with Western AI experts make it challenging to restrict AI advancements fully.

This development adds fuel to the debate over the release of open-source AI models, with some arguing it exposes sensitive technologies to misuse. Others advocate for continued transparency to foster global innovation despite security concerns.

Read more: Reuters

🧑‍💻 AI code

In Google’s Q3 2024 earnings call, CEO Sundar Pichai shared that AI now generates over 25% of the company’s new code.

Here’s why this is important:

• This shift, while aimed at accelerating development and innovation, brings new challenges for software engineers, especially those in entry-level roles.

• While AI can handle repetitive coding tasks, Pichai emphasized that engineers must now adapt by developing skills to supervise and refine AI-generated code, focusing more on complex problem-solving and innovation.

This push to embed AI into Google’s workflows highlights the growing role of AI in software development. To support faster AI deployment, Google has integrated its research, machine learning, and security teams.

Furthermore, Google’s Gemini model has expanded beyond Google’s ecosystem and is now available via GitHub Copilot, making advanced AI tools accessible to developers worldwide.

Looking ahead: The company is also investing in AI-driven video tools, with Google DeepMind’s Veo set to launch on YouTube Shorts.

Through these advancements, Google aims to take the lead in AI innovation, transform coding processes, and enhance product capabilities in the evolving tech landscape.

Read more: India Today

🖼️ Agentforce AI

Salesforce launched Agentforce, an AI-powered platform for developing chatbots and virtual agents, now available for general use.

Here’s the TL;DR:

• Agentforce is a low-code / no-code platform that allows businesses to deploy AI agents for customer and employee support, with clients like OpenTable, Saks, and Wiley already onboard.

• Agentforce’s AI agents operate autonomously, executing tasks based on data changes, rules, or pre-built automations, although Salesforce emphasizes the platform’s collaboration between human and AI agents to drive customer success.

• Agentforce includes features beyond traditional chatbots. It enables customizable agent creation through its Agent Builder, allowing businesses to tailor AI interactions using templates. It also offers Agentforce Service Agent, a self-service option priced at $2 per conversation.

• CEO Marc Benioff highlighted Agentforce’s competitive edge, referring to Microsoft's AI Copilot as “Clippy 2.0,” a jab at the reliability of Microsoft’s AI solutions.

Overall: This release highlights Salesforce’s deepening commitment to AI as it transforms its tools, including Slack, into what Slack CEO Denise Dresser calls a comprehensive “work operating system.”

Agentforce’s release marks Salesforce’s significant step forward in the enterprise AI race, positioning it as a key player in reshaping customer and employee engagement with AI-driven solutions.

Read more: TechCrunch

🛠️ AI Tools to Check Out

• MeetGeek: Automatically video record, transcribe, summarize, and
  share granular insights from every meeting to any tool. Check it out!

• Wikiwand: AI-driven wiki aggregator created to enhance user experience on Wikipedia by streamlining knowledge consumption. Check it out!

• Moonbeam: the only AI writing assistant specifically trained to help you write essays, stories, articles, blogs, and other long-form content. Check it out!

🤑 AI Fundraising News

• Decart raises $21M in funding to provide a platform that enables organizations to train massive AI models at scale while reducing training costs and improving efficiency.

• Persana AI raises $2.3M in funding to provide an app that leverages AI agents and multi-source data integration to automate GTM prospecting and engagement.

🐦 Tweet Post of the Day

An all-new AI designer 👇️