• Brain Bytes
  • Posts
  • OpenAI Supercharges ChatGPT Visuals, Apple Maps Fuels AI Growth, and Synthesia Shakes Up Actor Pay

OpenAI Supercharges ChatGPT Visuals, Apple Maps Fuels AI Growth, and Synthesia Shakes Up Actor Pay

30/03/25 - Brain Bytes

In partnership with

It’s not you, it’s your platform

Your newsletter content is top-notch. Your social media game is on point. Your aspirations are through the roof. Yet you’re stuck on level one.

We’ve got bad news (and then really good news): your newsletter platform stinks. And beehiiv can get you where you want to be.

Most newsletter platforms leave the growth and monetization aspects of content creation to the creator. But that’s like trying to drive a car with no wheels. 

Beehiiv’s platform is built for growth, so you can scale at the speed your content deserves with features like paid subscriptions, referral programs, and an ad network that connects you with global brands like Netflix and Nike.

It’s time to ask yourself: if your newsletter platform isn’t doing enough for you, what are you going to do about it?

1. OpenAI Enhances ChatGPT with Advanced Image Generation:

OpenAI has introduced GPT-4o, an upgraded version of its AI system, now capable of generating highly realistic images. This enhancement replaces the previous DALL-E 3 model and is accessible to all ChatGPT users across different subscription tiers. GPT-4o is a multimodal model adept at processing and generating text, video, audio, and images. Its development involved reinforcement learning from human feedback (RLHF), where over 100 human trainers provided input to refine the model’s accuracy. As a result, GPT-4o can produce lifelike images, coherent text, and even transparent backgrounds suitable for logos and presentations. Despite these advancements, the model occasionally exhibits inaccuracies, such as misrepresenting image elements. OpenAI asserts that GPT-4o respects artists’ rights by avoiding direct imitation of their work. Businesses, including GoDaddy, are already leveraging GPT-4o for AI-driven content creation, signaling a significant shift in digital content production.

2. Apple Utilizes Look Around Imagery to Enhance AI Training:

Apple has announced plans to leverage imagery and data collected from its Apple Maps Look Around feature to train its artificial intelligence models, starting in March 2025. Look Around, similar to Google's Street View, captures detailed images and 3D data through specialized vehicles and equipment. To ensure privacy, Apple blurs faces and license plates, and only anonymized data will be used for AI training. This initiative aims to improve various AI-powered services, including image recognition, creation, and enhancement. Potential applications encompass features like Image Playground, Clean Up tools in the Photos app, and advanced image search capabilities. Apple’s decision to utilize this extensive imagery dataset marks a significant step in enhancing the accuracy and effectiveness of AI tools within its ecosystem. Furthermore, this aligns with Apple's broader AI strategy, emphasizing responsible integration and privacy-conscious advancements in user experience, reinforcing the company's long-standing commitment to user privacy and ethical use of AI technology.

3. Krafton’s InZoi: Ambitious AI Integration in Life Simulation Gaming:

Krafton's latest venture, InZoi, is an ambitious photorealistic life simulation game integrating generative AI, released in early access on March 28, 2025. The game offers robust character creation and impressive environmental rendering. However, it currently faces significant performance challenges, such as asset pop-ins, character collision issues, and graphical glitches, particularly on mid-range hardware. The generative AI intended for item creation, clothing, and emotes exhibits inconsistent performance, highlighting the complexity and difficulty of effectively integrating AI into immersive gaming experiences. Krafton has acknowledged these early hurdles and has already rolled out patches aiming to address performance stability and AI-generated content inconsistencies. Despite these challenges, the company remains committed to refining and optimizing InZoi, demonstrating both the potential and limitations of AI in interactive entertainment.

4. Diffusion Models: Revolutionizing AI-Generated Content:

Diffusion models have emerged as a groundbreaking advancement in AI-driven content creation. These models function by introducing noise to training data, rendering it unrecognizable, and then learning to reverse this process to generate new, unique data. This approach enables the creation of diverse and complex outputs from simple inputs. The versatility of diffusion models lies in their ability to operate without structured training data, making them applicable across various domains, including art, music, science, architecture, fashion, and medical research. In the medical field, for instance, diffusion models enhance diagnostic imaging capabilities, aiding in more accurate assessments. However, challenges such as the need for substantial computational power and high-quality training data persist. Additionally, concerns regarding the potential misuse of diffusion models, including the creation of deepfakes and the amplification of biased data, underscore the necessity for ethical considerations as this technology continues to evolve. 

5. Synthesia Offers Equity to Actors for AI Avatar Contributions:

Synthesia, a London-based AI startup valued at $2.1 billion, is pioneering a novel approach to compensating actors who contribute to its digital “avatars.” The company has established a $1 million equity pool to reward actors with company shares in exchange for the use of their likenesses and collaboration in product development. Actors consent to a three-year usage period and receive a day’s pay for their participation, with the option to withdraw their appearance at any time. This initiative addresses growing concerns about fair compensation in creative industries, where artists often receive one-time fees while their work is used indefinitely. Synthesia’s move comes amid increased scrutiny over the use of AI-generated avatars in political propaganda, prompting the company to implement stricter moderation policies. By offering equity, Synthesia aims to align the interests of actors with the company’s growth, setting a precedent for ethical practices in the AI industry.  

6. OpenAI Faces Criticism Over AI-Generated Art Styles:

OpenAI has encountered backlash following the introduction of a feature that enables ChatGPT to generate images in the style of copyrighted works. Users have utilized this capability to create visuals mimicking the styles of renowned artists and characters, raising concerns about the unauthorized use of copyrighted material without compensation. Notably, Hayao Miyazaki, co-founder of Studio Ghibli, has expressed opposition to AI-generated art. In response to the controversy, OpenAI has restricted certain style requests to paying customers and limited imitations of living artists. Legal experts anticipate that this feature could lead to additional lawsuits against OpenAI. Advocates for stronger copyright protections argue that AI companies are exploiting artists’ work, calling for more stringent regulations to safeguard creative intellectual property. 

Ever wanted to create your own Newsletter? Beehiiv is the Shopify of Newsletters. Seriously, you should try it out. Click this image for 20% OFF your first 3 months.