Posted inGlobalNews

Horror of AI 2024: Removing the ‘Social’ Out Of ‘Social Media’?

Why does AI continue to scare humans?


Can AI Replace Humans?

The growth of generative AI content has been rapid, and it will continue to accelerate as more site administrators and publishers seek to enhance optimization and expedite production with advanced digital tools.

But what happens when AI content replaces human input? What happens to the internet when everything is simply a copy of a copy of a digital representation of actual human output?


That’s the issue many people are asking as social media platforms try to build walls around their datasets, leaving AI startups scurrying for new inputs for their LLMs.

It’s Been Happening …

X (previously Twitter), for example, has raised the price of API access in order to prevent AI platforms from using X postings while developing its own “Grok” model based on them. Meta has long restricted API access, particularly since the Cambridge Analytica scandal, and it also touts its unrivaled data pool to power its Llama LLM.

Google also signed an agreement with Reddit to incorporate its data into its Gemini AI systems, which is another route you can expect to see more of as social networks that aren’t building their own AI models look for new revenue streams through their insights.

The Wall Street Journal reported today that OpenAI contemplated training its GPT-5 model using publicly available YouTube transcripts, citing fears that demand for valuable training data will outweigh availability within two years.

It’s a big concern because, while the latest crop of AI tools can generate human-like language on almost any topic, it’s not yet “intelligence” in the traditional sense. Current AI models utilize machine logic and derivative assumptions to arrange words in order based on human-created examples in their database. However, these systems are incapable of thinking by themselves and are unaware of the significance of the data they generate. It’s advanced arithmetic in both text and graphic form, characterized by a systematic logic.

This means that, for the time being, LLMs and the AI tools based on them are not a replacement for human intellect.

That, of course, is the promise of “artificial general intelligence” (AGI), systems that can mimic human thought and generate their own logic and reasoning to complete certain tasks. Some argue that this is not far from becoming a reality, although the systems we can now access are nowhere near what AGI may theoretically do.

Other AI Concerns

Many AI doomers are also concerned that if we do develop a system that replicates a human brain, we would render ourselves obsolete, with a new, technological intelligence poised to take over and become the dominant species on Earth.

But most AI academics don’t believe that we’re close to that next breakthrough, despite what we’re seeing in the current wave of AI hype.

Meta’s Chief AI scientist Yann LeCun discussed this notion recently on the Lex Friedman podcast, noting that we’re not yet close to AGI for a number of reasons:

“The first is that there is a number of characteristics of intelligent behavior. For example, the capacity to understand the world, understand the physical world, the ability to remember and retrieve things, persistent memory, the ability to reason and the ability to plan. Those are four essential characteristic of intelligent systems or entities, humans, animals. LLMs can do none of those, or they can only do them in a very primitive way.”

According to LeCun, the amount of data that humans consume greatly exceeds the capabilities of LLMs, which rely on human insights gleaned from the Internet.

“We see a lot more information than we glean from language, and despite our intuition, most of what we learn and most of our knowledge is through our observation and interaction with the real world, not through language.”

In other words, the fundamental key to learning is interaction rather than language replication. In this sense, LLMs are advanced parrots capable of repeating what we have said to them. But there is no “brain” that can comprehend all of the human factors underlying that language.

Stay updated on all of the latest news by subscribing to the ITP Live newsletter below or by clicking on the push notifications.