OpenAI’s DIY AI Tools, AI Funding Surge, and Microsoft’s AI Model Independence

OpenAI has unveiled new tools empowering developers to create their own AI agents. Meanwhile, funding for AI startups has seen unprecedented growth, and Microsoft is leading the way in promoting model independence amidst mounting risks and policy discussions in the US.

Published on
March 19, 2025
8
min read
Article Image

⚡ Quick News

  • Billion-Dollar Deals & Valuations Major investments have propelled the AI sector, with Databricks, OpenAI, and xAI securing billions in funding. These deals contributed to the rise of 32 new AI unicorns in 2024.
  • Musk could still stop OpenAI’s for-profit transition OpenAI's move to a for-profit model continues, but Elon Musk might still sway the direction. Despite advancements in this transition, Musk's influence could pose a challenge.
  • Anthropic: China’s stealing $100M AI secrets! Dario Amodei, CEO of Anthropic, highlights concerns over industrial espionage targeting US tech firms by Chinese spies. He calls for increased US government intervention to safeguard sensitive AI data.
  • Judge Allows Authors' AI Copyright Case vs. Meta Authors have been granted permission to proceed with a copyright lawsuit against Meta's use of their works for AI training. The ruling, however, dismissed some additional claims in the lawsuit.

🛠️ OpenAI Unveils DIY AI Agent Tools

Main Story Image
OpenAI has launched groundbreaking tools designed to empower businesses to build customized AI agents, signifying a substantial move toward introducing autonomous AI systems within enterprises. The newly released Responses API allows companies to leverage the same technology powering OpenAI's Operator, combining capabilities like web search, file scanning, and software navigation. This API replaces the older Assistants API, scheduled to be deprecated by 2026. Alongside this, an open-source Agents SDK provides developers with tools to manage multi-agent systems, implement safety measures, and monitor performance closely. Notable early adopters like Stripe and Box have already utilized these resources, developing agents for invoicing and document searches, respectively.

Key Highlights:
  • The Responses API enables the creation of AI agents for web and document searches.
  • The Agents SDK is open-source, offering frameworks for secure multi-agent system integration.
  • Stripe and Box have successfully implemented these new tools for specific enterprise tasks.
  • OpenAI’s initiative emphasizes flexibility and customization in AI application development.
  • The initiative indicates a shift from the older Assistants API, which is set to retire in 2026.
Why It Matters: This launch signals a significant expansion in OpenAI’s toolkit, positioning businesses to develop tailored AI solutions that can operate autonomously within their unique environments. As 2025 is projected to be the year AI agents become more commonplace, these advancements could drive competitive advantage in various industries.

If you're enjoying Nerdic Download, please forward this article to a colleague. It helps us keep this content free.

💸 AI Funding Boom Reaches New Heights

Main Story Image
The fourth quarter of 2024 witnessed an unprecedented surge in AI funding, with global investments topping a record $43.8 billion, fueled primarily by large-scale mega-rounds and concentrated funding in key AI sectors. Venture capitalists directed nearly one-third of worldwide investments into AI companies, highlighting the field’s escalating prominence within tech finance. The quarter's investment was overwhelmingly driven by just 13 massive deals, each exceeding the billion-dollar mark, focusing on firms advancing AI models and infrastructure. This trend underscores a shift toward fewer, yet significantly larger, investments that prioritize industry leaders.

Key Highlights:
  • Q4 2024 saw AI funding peak at $43.8 billion globally.
  • A mere 13 mega-deals accounted for the majority of this investment rush.
  • Venture capital funding funneled 80% towards large, preeminent AI deals.
  • The quarter caps a year where private AI firms drew $100.4 billion in total funds.
  • This funding pattern underscores high confidence in prominent AI infrastructure providers.
Why It Matters: The concentration of substantial capital in few hands suggests a robust belief in the transformative potential of AI by the investment community. As AI continues to evolve, these funding patterns might not only accelerate innovation within the field but also increase market consolidation and power for top-tier companies, shaping the competitive landscape for years to come.

🔍 Microsoft Pioneers AI Model Independence

Main Story Image
Microsoft is pursuing AI model independence by developing its own in-house AI reasoning models, named MAI, to reduce reliance on OpenAI for its CoPilot suite of tools. These new models promise performance similar to existing industry leaders and are designed to replace OpenAI’s GPT within Microsoft's AI applications. As tensions rise between Microsoft and OpenAI regarding data sharing, Microsoft is also exploring partnerships with alternative AI providers such as Meta, xAI, DeepSeek, and Anthropic, seeking more autonomy in its AI deployments. Recent adjustments to their partnership have even allowed OpenAI access to diverse cloud services, which indicates strategic repositioning by both companies.

Key Highlights:
  • Microsoft is developing MAI, its new set of AI models, aimed at rivaling OpenAI's offerings.
  • Tensions arise from OpenAI’s reluctance to share technical data with Microsoft.
  • Microsoft considers AI models from other industry players to integrate into CoPilot.
  • The effort aligns with Microsoft’s strategy to reduce dependency on a single provider.
  • These changes follow a renegotiated contract allowing OpenAI more cloud provider flexibility.
Why It Matters: Microsoft’s initiative signifies a major strategic shift in how leading tech firms manage their AI capabilities, emphasizing self-sufficiency and diverse tech partnerships. This move not only diversifies Microsoft's technological base but also catalyzes further innovation and competitive dynamics within the AI ecosystem.

🏛️ US National AI Policy Input Accelerates Amid Risks Concerns

Main Story Image
The White House is rapidly soliciting feedback on its AI Action Plan, which is poised to bolster U.S. leadership in artificial intelligence while minimizing potential existential risks. Ordered in early 2025, this plan seeks to replicate the economic benefits seen with early nuclear energy realizations, avoiding historical pitfalls of public distrust witnessed during significant nuclear crises. The government is actively seeking opinions on agile regulation frameworks, crisis management coordination among tech companies and public entities, and the societal impacts stemming from AI-driven changes in labor dynamics. Discussions around accountability measures are also paramount, as the administration navigates global AI governance trends, particularly in contrast to the EU's regulatory challenges.

Key Highlights:
  • The plan encourages feedback on lightweight AI regulation models to enhance innovation.
  • Stakeholders are brainstormed on crisis response partnerships with tech firms.
  • Addressing AI-induced workforce disruptions forms a critical part of policy considerations.
  • Proposals highlight alignment with executive orders aimed at reducing development barriers.
  • Global AI governance pressures urge calibrated U.S. policy, particularly against China’s initiatives.
Why It Matters: The U.S. aims to remain at the forefront of AI development by balancing innovative thrusts with prudent governance strategies. By integrating public and private sector insights, the policy may set a precedent in effectively managing AI’s dual potential of economic prosperity and inherent risk, thereby sustaining U.S. competitiveness globally.

🛠️ New AI Tools

  • Clear Conscience Video Model by Moonvalley Moonvalley introduces Marey, a video model focused on producing quality AI content using exclusively clean and licensed data. This ensures authenticity, legal compliance, and high ethical standards in AI video creation.
  • Reka Flash 3: Advanced Reasoning AI Reka Flash 3 is an open-sourced 21 billion parameter AI designed for seamless on-device applications. It enhances on-device AI processing with substantial reasoning capabilities and efficiency.
  • Strategy Savant: Maestro by AI21 Labs AI21 Labs unveils Maestro, an AI agent adept at solving complex tasks through structured planning and strategic analysis. It optimizes resource allocation by evaluating strategies, costs, and outcomes.
  • IBM's Granite 3.2 Reasoning Model IBM introduces Granite 3.2, an open-source model that alternates between standard and chain-of-thought reasoning. This model enhances AI problem-solving akin to the approach used in Claude 3.7.