OpenAI's bold moves redefining the future of innovation

NOVEMBER 13 2023

Hi there,

this week we discuss OpenAI's game-changing announcements. We also cover how to create video game characters using DALLE-3, Elon Musk's 'Grok' AI bot, Amazon's investments in large language models, the UK's AI supercomputer project, and more. Did you know that training a single AI model consumes more electricity than 100 U.S. households in a year? We can help optimize AI algorithms, reducing their reliance on cloud infrastructure.

AI top story

👏 OpenAI's bold moves redefining the future of innovation

OpenAI 's groundbreaking announcements are transforming the AI landscape, ushering in a new era of possibilities for developers, startups, and businesses. With the introduction of the Assistants API, developers can now create intelligent AI agents for their applications effortlessly. These agents are equipped with specialized functions, including Code Interpreter, Retrieval, and function calling, streamlining task execution. The days of complex Retrieval Augmented Generation (RAG) pipelines are over, as users can seamlessly expand GPT's knowledge by uploading files, eliminating the need for GPT wrappers like ChatPDF and LangChain .

Moreover, the Assistants API introduces persistent threads, enabling developers to manage extended conversations and complex tasks, fostering more coherent and context-aware interactions. Developers gain the ability to run Python code, handle diverse data, create visual content, and access external knowledge sources without the hassle of embedding or extensive dataset searches. Custom functions can be defined and called via the API, broadening the range of achievable tasks. Alongside this, GPT-4 Turbo emerges as a cost-effective, high-speed version of GPT-4 with a 128k context window, accommodating over 300 pages of text in a single prompt. It offers GPT-4-level capabilities at a 2.75x reduced cost, unlocking previously unviable applications and ensuring reduced latency for user-facing apps.

OpenAI's developments position it as the potential foundation for AI, yet open source will remain a preference for organizations seeking customization and cost control.

Wonder what's best for you? Reach out to Ivan Draganov .

Latest news

📚What we’ve been reading

1. Elon Musk debuts ‘Grok’ AI bot to rival ChatGPT, others (1-minute read)
2. Amazon is investing millions in training ambitious LLMs, hoping it could rival top models from OpenAI and Alphabet (1-munute read)
3. The UK is spending $273 million to build its fastest ever AI supercomputer (2-minute read)
4. An Efficient Self-Supervised Cross-View Training For Sentence Embedding (20-minute read)
5. Are language models good at making predictions? (5-minute read)

Tips and tricks

🎮 Create characters for a video game using DALLE-3

When I was younger, I liked playing Valve corporation 's Counter-Strike. So, I wanted to create some additional characters in the game - a mechanic, a doctor, a character that has a dog helper, and two additional female characters to make the game female-friendly. Here's a breakdown of the steps:

1. First, conceptualize your character. Begin by defining the character's role, personality, and backstory within the game's narrative. Describe the character's physical appearance, including details like age, gender, height, build, and any unique features such as scars, tattoos, or accessories.
2. Write a prompt to DALLE-3 to create the character. Iterate, if necessary. Once satisfied, ask DALL·E-3 for the image ID associated with the character for reference.
3. Conceptualize a new character using the tips in 1) and use the image ID to keep consistent colors or features as you create the other characters.See the final results here.

🔢 This week's number: 100

Training a solitary AI model or chatbot consumes more electrical energy than the annual usage of 100 typical U.S. households. For instance, the training of ChatGPT alone devoured approximately 1.287 gigawatt-hours of energy, which roughly equals the yearly consumption of 120 American homes.

💰Save up to $10,000 with a tailored Gap Analysis

Do you struggle to assess the potential return on investment (ROI) and conduct a thorough cost-benefit analysis?

A Tailored Gap Analysis by LyRise is a process that helps businesses understand how to make the most out of Artificial Intelligence (AI) for their business.

To get your Free Gap Analysis, reach out to Ivan Draganov, who will guide you through the process.