AI Services Are Everywhere, Ushering in the Return of Subscription Fatigue

  • Numerous platforms are vying for users who wish to test out and benefit from AI.

  • However, using such services comes with a high cost, and subscription models are becoming increasingly prevalent.

  • This trend is contributing to the saturation of subscription-based services, which appears to be approaching a breaking point.

No comments Twitter Flipboard E-mail

Need to remove a background from your image? No problem, AI can do it for you in the blink of an eye. Want to create the first draft of your design in record time? There are several services for that. Need AI to write a text that's perfect and creative for you? There are similar solutions for that, as well. Generative AI models have become incredibly useful tools across various fields. However, they’ve also intensified existing issues.

Try it for free... Although most of these services offer free trials, there are usually some restrictions. For instance, you might be able to generate an image, transcribe audio, or summarize a PDF using AI, among many other things. But if you need to do these tasks multiple times, you'll need to pay for the extra usage.

Get a subscription… If the free trial hook worked, you might want to pay for the service's monthly subscription plan for unlimited access. A good example of this is ChatGPT and ChatGPT Plus. They’re almost the same but cater to occasional users and heavy users. With the free version, you’re limited in the number of queries you can make, and you can’t generate images. ChatGPT is essentially a way to entice you to sign up for ChatGPT Plus, which costs $20 per month.

…or buy a token. Another option is to buy credits instead of subscribing, which allows you to use the service in full. That's how services like Magnific, an AI creative upscaling app, work. Using the service consumes tokens based on the resolution of the images you’re working with. When you run out of tokens, you can purchase more to keep using the service. At the time of publication, you could get 2,500 tokens for normal upscales (about 200 images) or large upscales (about 100 images) for $39 per month.

Using AI can be expensive. And not just because of the subscription costs, but because the underlying infrastructure is actually very expensive. Platforms offering various AI options are built on large language models (LLMs) and AI models from major companies. These platforms use these models through APIs to provide tools for specific purposes. However, access to data centers running these processes requires costly GPUs, which also consume significant amounts of energy, leading to additional expenses. Ultimately, users bear these costs if they want continuous access to these services. Additionally, the effectiveness of the model, as recognized by Adobe itself, adds to the appeal.

Subscription fatigue is making a comeback. With the overwhelming number of content streaming platforms and various services that rely on subscription models, we’re constantly bombarded with monthly fees. This issue also extends to AI services, which are contributing to the problem of subscription fatigue. The real question is: How much longer can we endure this?

Too many players. Currently, the market seems quite diverse. Apart from a few major players like ChatGPT and Midjourney, subscriptions to AI platforms are doing well. However, there may be a trend towards consolidation in the future. The development of platforms like ChatGPT, Gemini, and Copilot indicates a future where these models, along with others, might handle a wide range of tasks, potentially leaving less space for more specialized services.

Image | Mia34s_64998 using Midjourney

Related | ‘Have I Been Trained?’ Is a Site That Helps You Find Out if Your Data and Work Have Been Used to Train AI

Home o Index