Why You STILL Won't Pay for AI and AI Assistants Anytime Soon

Today let's just jump right into it. 2024 has been nothing short of a massive year for LLM AI's for text-to-text, text-to-image and right at the buzzer in December text-to-video. In December OpenAI had their false start launch with SORA which temporarily broke ChatGPT taking it offline around the world, and Google has released Gemini 2.0 and its beta release of VEO 2 - a video model which looks so incredible SORA already looks obsolete.

With the release of all this wonderful AI tech there are also some trends that were greenshoots in early to mid 2024 that are now becoming much clearer, specifically around the two things we wrote to our subscribers about in June 2024.

In June 2024 we wrote a post titled "You won't pay (directly) for AI" and in it we explained how " What Apple's integration of OpenAI within Apple Intelligence really demonstrated is that consumers paying directly for AI isn't going to be a thing anytime soon.

What's fast becoming obvious is getting AI usage into the hands of billions of people in order for the LLMs to use and learn from their behaviour, is worth far more (currently) than resricting access or keeping them behind a paywall."

We also wrote to subscribers talking to them about how with tech giants scrambling for users and usage rather than subscriber fees AI Assistant's are basically coming to everything "Assistant Intelligence is coming to everything".

Why You Won't Need to Pay for LLMs Anytime Soon

AI is a hype-playground and tech companies are vying for your attention, but more important than consumers opening their wallet and paying is usage, because usage creates new data, which is desperately needed to feed the intelligence and optimisation to train new AI models.

Twitter, which made their AI Assistant Grok free and open to everyone this week, and OpenAI, which frequently makes AI models free to use that were previously paid access only, are doing so in a bid to expand their user base to fuel company growth.

Others, like Google, Facebook and Microsoft, are playing both offense and defence. They want to attract new users while simultaneously protecting their existing dominance from AI upstarts.

For all of these giants, the value of getting their latest AI in consumers hands and professional creatives to get the volume of usage into their models they need to fuel the next generation of AI models outweighs the immediate financial gain from subscription fees.

New releases in large language models (LLMs), video and image generation capabilities showcase leadership and attract both users and investors. Early access translates into valuable market data – how you interact with the AI, its strengths and weaknesses – which fuels further development. That information, not your subscription dollars, is the real prize.

The Money Game: Who Needs Your Cash?

Giants like Google don't desperately need additional revenue streams from AI. Google and Microsoft both boast huge and diverse income portfolios with established products like advertising that fund their AI pursuits. This allows them to focus on innovation without being pressured by immediate financial returns for specific AI products.

OpenAI and Twitter, on the other hand, doesn't have this luxury. They require user adoption and revenue growth to survive.

While they might offer paid tiers with exclusive features, the core functionality might remain free, fuelled by user data and partnerships.

The Holy Trinity of AI Domination

Our analysis places Google and Microsoft in the pole position of the AI race. They possess the "holy trinity":

1. Financial Muscle: They can afford the best researchers and development teams.
2. Self-Funding AI: They don't rely on direct AI revenue to keep advancing.
3. Massive User Bases: From laptops and PCs to tablets and mobiles, they already have you plugged into their ecosystem, providing rich user data for AI optimisation.

This trinity grants Google and Microsoft in particular a significant advantage in the long run.

The Pace of AI Creates a Unique Paradox

The breakneck speed of LLM and 'text-to-n' AI development presents a unique paradox: it may ultimately stifle the growth of the very industry driving it. As AI capabilities rapidly expand, offering increasingly sophisticated solutions for free, the need for paid AI services diminishes. This phenomenon, where AI advancements cannibalize the market for premium AI tools, poses a significant challenge to the long-term sustainability of the AI industry.

The core issue lies in the accelerating pace of AI innovation. Every few months, new models emerge with enhanced capabilities, often surpassing the previous generation in performance and accessibility. This rapid progress leads to a situation where yesterday's cutting-edge AI solution becomes today's commodity, quickly integrated into free platforms or open-source projects.

Consider the evolution of image generation. Initially, high-quality AI-generated images were largely confined to paid subscription services. However, with the rapid advancement of open-source models and the democratization of access to powerful AI tools, generating impressive images has become increasingly accessible to the general public for free. This trend is not limited to image generation; it's observable across various AI domains, from text generation and code completion to data analysis and even basic research.

As a result, users are becoming increasingly satisfied with the capabilities of free AI tools. The perceived value proposition of paying for premium AI services diminishes as readily available alternatives offer comparable or even superior performance. This shift in user behavior can have a profound impact on the AI market, potentially eroding revenue streams for companies heavily reliant on paid subscriptions.

This phenomenon, where AI advancements cannibalize the market for premium AI tools, is not merely a theoretical concern. It's already evident in the evolving landscape of AI services. Many companies are exploring alternative revenue models, such as offering premium features on top of free basic services or focusing on niche applications with specialized needs.

The pace of AI change, while exciting, presents a unique set of challenges for the industry. Companies must adapt to this dynamic environment, finding innovative ways to monetize their AI offerings while navigating the ever-shifting landscape of free and open-source AI tools.

Apple's Privacy Problem

But what of Apple? As a premium provider of technology devices with strong privacy controls in place Apple's position in the AI race is curious. They possess strong AI desire, evident in their partnership with OpenAI for a more intelligent Siri. However, their stringent privacy stance creates a dilemma.

Maintaining their privacy focus is almost sure to hinder their ability to deploy the cutting edge AI products and services that, while not immediately impactful now, could become a real blocker to future ambitions.

Many of Apple's on-device AI processes are clunky, cumbersome and we imagine users won't tolerate them for long. For those that don't know, in some instances where a person wants their phone to us AI, Apple asks the user if Apple can send the users data off device to complete a certain task.

Compared to non-Apple devices, this experience is demonstrably clunkier and worse. Users won't tolerate long wait times for AI tasks when other devices complete them faster and more comprehensively. 

By consistently presenting iPhone and iPad users with privacy choices, we believe they'll often prioritise task completion. This could desensitise them to minor data sharing, potentially eroding overall privacy concerns. This approach may also empower Apple to shift default privacy settings or introduce global privacy controls, enabling users to manage their privacy more comprehensively on their devices.

Apple faces a crucial decision:

• Uphold Privacy: This might result in falling behind the AI curve and a less competitive Siri.
• Loosen Privacy Reins: This could risk user trust and chip away at their privacy-focused brand image.

Our view? If they don't adapt soon, their AI efforts may falter, and Siri will remain an underwhelming companion. Apple can't afford to miss the AI revolution entirely.

The Bottom Line: Free AI Today, Paid Solutions Tomorrow?

So, will you be paying directly for AI anytime soon? Probably not. For the forseeable future the big tech companies will prioritise user adoption and data collection above immediate financial gain. They see user data as the most valuable resource – user interactions with their AI inform their development roadmap.

However, this doesn't mean the future is free of charge. As AI matures, specialised, premium features in industries such as realestate, medicine, health and others might emerge with a price tag. Think of advanced creative tools or highly customised business AI solutions.

Let's see how this game unfolds and what the future holds for AI monetisation, and enjoy it while it is free.