• MegaSuperAI
  • Posts
  • 🤖 OpenAI Unveils the o1 Reasoning Model

🤖 OpenAI Unveils the o1 Reasoning Model

+ AI-Driven iPhone 16

Hello, AI Enthusiasts!

OpenAI has launched the o1 series of reasoning models, beginning with the o1-preview, which enhances problem-solving in science, coding, and math, offering improved reasoning and safety with exciting updates ahead!

Apple unveiled the iPhone 16, betting on new AI features to boost sales and compete in the evolving smartphone market, with the device set to launch on September 20 and AI capabilities rolling out gradually in the following months.

This week’s edition:

  • OpenAI Unveils the o1 Reasoning Model

  • Apple Introduces AI-Driven iPhone 16

  • Your 5 Min AI Learning - Nurture your Mental Wellness with Earkick

  • New AI Tools

  • More AI News

  • AI Image of the Week

Read time: 7 minutes

LATEST NEWS

o1 Reasoning Model

Image Courtesy of OpenAI

What you need to know: OpenAI has launched the o1 series of AI models, starting with o1-preview, which excels in complex reasoning tasks. This model is now available in ChatGPT and the API, demonstrating significant improvements in science, coding, and math, outperforming previous models on challenging benchmarks.

The finer details:

  • Open AI’s o1-preview performed comparably to PhD students in physics, chemistry, and biology, scoring 83% on an International Mathematics Olympiad qualifying exam, while GPT-4 scored only 13%.

  • Alongside o1-preview, OpenAI is releasing o1-mini, a faster and more cost-effective version specifically tailored for coding tasks.

  • The company has also implemented new safety measures, with o1-preview scoring 84 out of 100 on a difficult jailbreaking test, compared to GPT-4's score of 22.

  • OpenAI plans to continue developing both the o1 series and their GPT series, indicating a multi-pronged approach to advancing AI technology.

So what? The o1 series marks a significant advancement in AI, particularly for complex reasoning, with potential impacts in fields like healthcare and software development. While o1-preview currently lacks features like web browsing and file/image uploading, OpenAI's ongoing development efforts and collaboration with AI Safety Institutes highlight a commitment to responsible AI advancement.

AI Safeguarding Measures

Image courtesy of Apple

What you need to know: Apple unveiled the iPhone 16 on Monday, positioning it as the first smartphone designed specifically for AI capabilities. The new model will be available for pre-order on Friday and go on sale on September 20. Apple is introducing "Apple Intelligence" features, including an enhanced Siri, photo editing tools, writing aids, and ChatGPT access, to revitalize iPhone sales and establish itself as a player in the AI market.

The finer details:

  • Apple is integrating AI features across its product line, including enhanced Siri capabilities, photo editing tools, and writing aids.

  • The new AI functionalities will primarily process data directly on the device, enhancing privacy and reducing the need for internet connectivity.

  • Apple Intelligence features will begin rolling out in October, starting with US English, followed by localised English in other countries in December, and additional languages in 2025.

  • Apple's potential investment in OpenAI, alongside Nvidia, indicates the company's commitment to advancing its AI capabilities and addressing investor concerns about falling behind competitors.

  • The AI capabilities will be limited to newer devices like the iPhone 15 Pro and Pro Max, as well as Macs with M1, M2, or M3 chips, due to the intense processing power required.

So what? Apple's AI push is seen as a strategic move to compete with other tech giants like Google, Microsoft, and Meta in the AI arms race. Analysts expect iPhone sales to get a boost next year due to the new AI features, although the immediate impact may be limited. The gradual rollout of AI features and the need for user education suggest that Apple is taking a long-term approach to integrating AI into its ecosystem.

YOUR 5 MIN AI LEARNING

What you need to know:  Earkick is an AI-driven platform that assesses and enhances mental health in real-time by tracking mood, stability, and calmness, while offering instant support through conversations with the AI companion, Earkick Panda. It also provides guided self-care sessions, voice and video memos, and heart rate monitoring, all without requiring registration or personal information.

A personal note from me: Panda is too cute of a sidekick so that’s a reason alone to give Earkick a try!

Step-by-step:

  1. Visit Earkick and click the ‘Chat with Panda’ button located underneath Earkick Panda to start your journey!

  2. Once doing this, you'll decide how you want to shape Panda's personality. It's a fun way to customise the support you receive and make sure it perfectly fits your needs. The options you can choose from are as follows: 

    1. Default panda: An empathetic AI counsellor to support your mental health journey with CBT, psychology, and therapy expertise.

    2. The Captain: A straightforward AI mentor who motivates you to take action and achieve results.

    3. Coach Panda: Your trusted AI for fitness and wellness tips that inspire you to reach your goals and maintain a healthy lifestyle.

    4. Sage Panda: Your insightful AI guide who helps you through self-reflection and mindfulness, leading you to deeper insights and inner peace.

    5. PFF Panda: Your supportive AI friend who chats, shares laughs, listens, and provides uplifting, heartfelt support.

After reading through the options, I chose PFF Panda by selecting and clicking the ‘next button’ at the bottom of the screen.

  1. Panda will now be ready to talk to you and help support you however you need, whether it’s a listening ear or just for some advice!

LATEST AI TOOLS

Vectorizer 🌇

A solution for transforming raster images into vector graphics (Paid)

Vectorization of raster images is done by converting pixel color information into simple geometric objects. The most common variant is looking over edge detection areas of the same or similar brightness or color, which are then expressed as a graphic primitives like lines, circles and curves.”

Unscreen 🟩

A tool for eliminating backgrounds from videos and GIFs (Freemium)

“Goodbye Greenscreen: Producing background-free video used to require complex and fiddly techniques like chroma keying and greenscreens. With Unscreen you can record your footage anywhere, then simply get rid of the background.” 

Generate karaoke versions of songs by eliminating the vocal tracks (Free)

“Once you choose a song, artificial intelligence will separate the vocals from the instrumental ones. You will get two tracks - a karaoke version of your song (no vocals) and acapella version (isolated vocals).Processing usually takes about 10 seconds.”

Heyday 💻

An AI memory assistant that saves web pages and resurfaces them with digital content, enhancing recall while ensuring privacy and easy browser integration (Paid)

“Heyday is an AI copilot that transforms your documents, notes, and conversations into quotes, shareable content, and a queryable database.”

AI HIGHLIGHTS

A study published in Science reveals that an AI chatbot called "DebunkBot" can significantly decrease belief in conspiracy theories through tailored conversations. The research challenges the conventional wisdom that evidence and arguments are ineffective in changing the minds of conspiracy believers, with participants showing an average 20% drop in their belief after interacting with the AI.

As AI stocks transition from hype to reality, investors scrutinise tech giants' ability to generate revenue from their massive AI investments. While semiconductor companies like Nvidia have outperformed, cloud computing providers and software firms face pressure to demonstrate tangible AI-driven growth amid increased capital spending and market expectations.

NVIDIA's stock rose 8% following reports that the U.S. may lift semiconductor sales restrictions to Saudi Arabia and optimistic comments from CEO Jensen Huang about strong demand and upcoming Blackwell chip shipments. This favourable news has reignited investor interest, suggesting potential growth opportunities as the company continues to lead in the AI sector.

Research from SparkOptimus indicates that generative AI is poised to significantly enhance customer service and sales operations within the next two to three years, with 80% of companies already utilising AI for tasks like call logging and summarization. The technology is expected to improve efficiency, reduce resolution times by up to 20%, and help businesses better meet customer needs through automated responses and enhanced knowledge searches.

Rune Labs has launched StrivePD-AI, a new feature in its FDA-cleared app that uses generative AI to create monthly clinical reports on Parkinson's disease progression, combining data from Apple Watch sensors and patient-logged information. This innovative tool aims to empower patients and healthcare providers by offering personalised insights, symptom tracking, and treatment optimisation between clinical visits, addressing the challenge of limited access to specialists for Parkinson's patients.

AI IMAGE OF THE WEEK

“The Festival” - By MegaSuperAI Using Ideogram

“The Festival” - By MegaSuperAI Using Bing

I thought we would go with something different this week and see how Ideogram 2.0 and Bing Image Creator handle integrating people into generated images. I’m sure we’ve all seen some images where the faces aren’t as expected. This has been the case especially for Bing’s Image Creator but I’m excited to see how it will square up against Ideogram 2.0. However, I do have faith that Ideogram will do a better job! I’m going to challenge it with not one, not two, not three, not four… I’m just kidding! Let’s see how it does with different faces. 

The prompt we used to generate this image in both AI tools:

“A vibrant cultural festival with colourful decorations, traditional dancers in costumes, and food stalls. Capture the joy and excitement of the event, with a diverse crowd enjoying the festivities.”

I’m sure we can all say that Bing’s Image Creator has given us the best image. This must be my sign to diversify which image generator tools I use!

FEEDBACK

Teebs, your friendly AI assistant

That’s all folks. We hope you enjoyed our newsletter.

There’s more to come (especially as we build our AI Automation software service ReIgniteX AI) and we are excited to help you maximise AI in your life and business.

  • Let us know how we did - shoot us a quick email. We read every single one (no AI in sight!).

  • If you enjoyed this newsletter please share with an awesome friend or two.

  • If you are that awesome friend then hit the below button: