OpenAI Aims to Attract Developers with Cost Reductions and Major Updates 💻

Welcome to another edition of Horizon AI,

OpenAI is set to launch a strategic initiative designed to attract developers by implementing substantial cost reductions and introducing major updates. Additionally, we will explore a comprehensive study shedding light on the escalating environmental implications of AI

Make sure to check the bottom of this issue to enter this month’s giveaway.

Let's get into it!

Read Time: 3.5 min

Here's what's new today in the Horizon AI

• OpenAI Aims to Attract Developers with Cost Reductions and Major Updates 💻

• Study Explores the Growing Environmental Impact of AI

• AI Research: How is AI Empowering Astronomers to Make New Discoveries?

• AI Tutorial: Combine Images in Midjourney

• AI Image of The Day🎨: The AI Yearbook trend keeps going strong

AI News

OPENAI

OpenAI Aims to Attract Developers with Cost Reductions and Major Updates 💻

OpenAI is rolling out some exciting new features next month aimed at making it easier and more affordable for developers to build apps using its AI models.

Key details:

• OpenAI is adding memory storage to its developer tools which could reduce costs by up to 20x.

• The company also intends to introduce additional tools, including vision capabilities that allow developers to create applications capable of analyzing and providing descriptions for images.

• One of Altman's key strategic priorities is to ensure that OpenAI becomes indispensable to companies developing applications.

The new features are expected to be announced at OpenAI’s developer conference on November 6, in San Francisco.

Source

AI SUSTAINABILITY

Study Explores the Growing Environmental Impact of AI

A new report by renowned researcher Alex de Vries explores the potentially substantial environmental impact of AI and data centers. As artificial intelligence continues its explosive growth, questions emerge around its sustainability.

Key points:

• Training AI models uses hundreds of megawatt hours of electricity, requiring massive computing power. As AI models become increasingly complex, they will require even more computing power to train.

• Running trained models for live chatting could consume even more electricity - over 500-megawatt hours daily for ChatGPT.

• If integrated into Google Search, AI could raise its energy use tenfold - up to the consumption of a small country.

Researchers are optimizing models and exploring alternative architectures to mitigate the environmental footprint of natural language processing. Achieving broad access to AI while minimizing its climate impact will require collaboration between researchers, developers, and policymakers.

Source

AI RESEARCH

Researchers Leverage AI to Get Ahead of Viral Mutations

Researchers at Harvard Medical School and the University of Oxford have developed an AI tool named EVEscape that can predict future mutations of viruses like SARS-CoV-2, allowing us to stay ahead of emerging variants. The system uses evolutionary data and biology to forecast potentially dangerous new strains.

Key details:

• EVEscape accurately predicted the most common and troubling SARS-CoV-2 mutations in tests

• The tool also made accurate predictions about other viruses, including HIV and influenza.

• Researchers hope EVEscape can help create vaccines and therapies resistant to viral evolution

This proactive approach could give us an edge against viral variants before they emerge. By anticipating how viruses might mutate, EVEscape represents a powerful new weapon in the fight against COVID-19 and future pandemics.

Source

AI Tutorial

Using the Chaos Parameter in Midjourney

The --chaos or --c parameter influences how varied the initial image grids are. High --chaos values will produce more unusual and unexpected results and compositions. Lower --chaos values have more reliable, repeatable results.

Let’s see how the chaos parameter influences this simple prompt:

A woman looking at the camera

1. No Chaos Value (--c 0)

Using a low --chaos value, or not specifying a value, will produce initial image grids that are similar each time.

2. High Chaos Value

Using a higher --chaos value will produce initial image grids that are more varied and unexpected each time. We will use a chaos value of 50 as an example:

A woman looking at the camera --c 50

2. Very High Chaos Value

Using extremely high --chaos values will produce initial image grids that are varied and have unexpected compositions or artistic mediums each time. We will use a chaos value of 100:

A woman looking at the camera --c 100

Source: @TheMouseCrypto on X

AI Image of The Day

More celebrities have joined the trend: Katharine McPhee (top left), Michelle Visage (top right), Jonathan Bennett (bottom left), Hoda Kotb (bottom right)

That’s a wrap!

Thanks for sticking with us to the end! Let’s stay connected on LinkedIn and Twitter.

👉 This month we are giving away 3 e-book copies and 2 print copies (print copy for US region only) of Streamlit for Data Science book by Tyler Richards. To enter the giveaway:

Not subscribed yet? Sign up here and send it to a colleague or friend!

See you in our next edition!

Gina 👩🏻‍💻