OpenAI nabs largest venture round ever

ALSO: Spotlight on Lended AI

Welcome to another edition of MerlinsNotes!

Here’s what’s on the desk this week:

  • OpenAI's massive funding round and DevDay announcements reshape the AI landscape

  • KKR's RealHouse platform

  • Lended AI emerges as a game-changing tool for independent sponsors & lenders

Let’s get into it.

FIRST PASS

OpenAI secures $6.6B funding at $157B valuation, unveils major upgrades at DevDay (CNBC)

OpenAI recently announced a massive $6.6 billion funding round at an eye-popping $157 billion post-money valuation.

The funding comes alongside significant announcements at their first-ever developer conference, DevDay, including major upgrades to their AI models and services, and a new Custom Models program.

Key points:

  1. The round was led by strategic investors including Microsoft, Nvidia, SoftBank, and others, cementing OpenAI's position as the dominant AI powerhouse.

  2. New Realtime API for building fast speech-to-speech experiences.

  3. Launch of the Custom Models program for select organizations.

  4. New Assistants API for building more complex, agent-like AI applications.

  5. Significant price reductions for API calls, making AI more accessible for developers.

Merlin’s Notes: OpenAI's latest moves are set to accelerate AI adoption across industries, with particular implications for roles involving voice—think outsourced call centers and customer service operations.

The ability to create custom AI assistants and even custom-trained models could revolutionize how businesses handle customer interactions, potentially leading to significant cost savings and improved customer experiences.

The new Custom Models program is particularly noteworthy. It allows selected organizations to work directly with OpenAI researchers to train custom GPT-4 models for specific domains, potentially posing a threat to large consultancies currently doing this work.

This includes modifying every step of the model training process, from domain-specific pre-training to custom reinforcement learning post-training.

This level of customization is especially valuable for industries with large proprietary datasets (billions of tokens at minimum).

With the Realtime API and other AI voice solutions, businesses can now create more natural, real-time voice assistants with ease, thus reducing the need for overseas agents that can come with their own challenges.

For PE firms, these developments present both opportunities and challenges:

  1. Portfolio companies in customer-facing industries may need to rapidly adapt to remain competitive.

  2. There's significant value creation potential in implementing these AI solutions to streamline operations and enhance customer engagement.

  3. The Custom Models program opens up new possibilities for companies with vast proprietary datasets to create truly unique AI solutions.

  4. Firms should consider how these AI advancements might disrupt traditional business models and create new investment opportunities.

As the AI landscape evolves, PE firms should consider:

  1. Assessing the AI readiness of portfolio companies, particularly those with large proprietary datasets.

  2. Identifying opportunities for AI-driven value creation, especially in areas like customer service and data analysis.

  3. Preparing for potential disruptions in labor-intensive industries, especially those relying heavily on knowledge workers or customer service operations.

JARGON BUSTER

Context Window: In AI language models, the context window refers to the amount of text the model can consider at one time when generating or processing language.

It's like the AI's short-term memory or field of vision for text.

Here's how it works:

1. Input: The user provides text input (e.g., a question or prompt).
2. Processing: The AI considers this input along with a certain amount of preceding text.
3. Output: The AI generates a response based on this "window" of context.

Think of the context window as a sliding glass door on a long text document. The AI can only "see" and work with the text visible through this door at any given moment.

USE CASE SPOTLIGHT

KKR’s RE analysis platform, RealHouse

Source: Insider

KKR, a leading global investment firm, has developed RealHouse, a new data science tool to enhance its real estate investment capabilities and analyze its portfolio of 1,000+ investments.

Key features of RealHouse:

  • Provides instant access to deal overviews, performance metrics, and related documentation.

  • Displays fund composition and risk exposures in one centralized platform.

  • Automates routine tasks like monitoring interest-rate caps and generating funding projections.

  • Enables faster decision-making by reducing time spent on data gathering and analysis.

Lead Executives:

  • Collaboration between KKR's technology and real estate teams.

  • Led by Jessica Ciaccia who worked closely with Steve Lo and Jon Knehr, directors on the technology team, to build the platform.

  • The team plans to expand the tool to the firm's real estate private equity business.

Impacts and Future Plans:

KKR's RealHouse platform has already made a significant impact on the company's operations:

  • Over 200 KKR employees across various departments are using RealHouse.

  • The platform provides a unified view of deal overviews, performance metrics, fund composition, risk exposure, and related documentation.

Growth and Strategy:

KKR's real estate credit business has experienced remarkable growth:

  • Expanded from $1 billion to over $37 billion in Assets Under Management (AUM) in less than a decade.

  • This initiative underscores the value of leveraging data for actionable portfolio insights.

  • Validates the shift towards building in-house software and data tools rather than relying on per-seat SaaS solutions.

Industry Implications:

The RealHouse platform demonstrates several key trends in the industry:

  • Growing importance of data science and technology in investing.

  • Necessity of strategic technology advisors within PE.

  • Potential for advanced analytics and AI to provide a competitive edge in investments.

  • Possibility of more efficient portfolio management and better-informed investment strategies.

TOOL OF THE WEEK

Lended AI

Source: Lended

Came across this AI take on the SBA loan process (H/T: @carrynointerest) and just had to share.

LendedAI is an innovative platform that's revolutionizing the SBA 7(a) loan application process, promising to deliver term sheets in just 24 hours instead of weeks.

The tool streamlines the often complex and time-consuming SBA loan application process through a combination of AI-powered assistance, expert guidance, and a user-friendly interface.

Here's what makes them stand out:

  1. Rapid turnaround: Get term sheets for your SBA 7(a) loan in 24 hours

  2. AI-driven matching: Intelligent algorithms match you with suitable lenders

  3. Streamlined application: One application connects you to 40+ lenders

  4. Expert support: Dedicated guidance throughout the process

  5. Cost-effective: The service works on a success-fee model, aligning their interests with yours

LendedAI breaks down the application process into manageable steps, including pre-check, loan application, lender matching, and term sheet generation.

Then, the platform's AI assesses your application and documents to find the best possible loan options, potentially saving business owners significant time and effort.

Here are their stats today:

  1.  40+ banks are on platform

  2. ~$50M in M&A deals underwritten

  3. Now expanding into working capital

  4. Banks say it helps accelerate their go/no-go evaluation by 90%


Have a question about AI? Or need help with a specific scenario?

We’ve introduced Ask Merlin to answer all your burning questions!

We’ll read every single question and use it to inform future editions. Submit your question by clicking on the button below.

Until next time! I’d welcome your feedback on today’s edition.

Hit reply to this email or shoot me a note at [email protected]

Thoughts on today's email?

No hard feelings either way!

Login or Subscribe to participate in polls.

May the tailwinds be ever in your favor,

— James

P.S. Did someone forward this email to you? If so, you can subscribe here!