{"id":15326,"date":"2026-04-24T11:02:09","date_gmt":"2026-04-24T11:02:09","guid":{"rendered":"https:\/\/www.europesays.com\/ai\/15326\/"},"modified":"2026-04-24T11:02:09","modified_gmt":"2026-04-24T11:02:09","slug":"ai-wont-steal-your-job-until-it-evolves-into-this-nightmare-personal-finance-news","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/ai\/15326\/","title":{"rendered":"AI won’t steal your job… Until it evolves into THIS nightmare | Personal Finance News"},"content":{"rendered":"

New Delhi: The rise of artificial intelligence has triggered large scale fears about job losses. But according to Vishal Mishra of Columbia University, current AI systems like Large Language Models (LLMs) are not as big a threat to jobs as many believe\u2014at least not yet.<\/p>\n

LLMs Are Powerful, But Constrained<\/p>\n

LLMs such as ChatGPT and other generative AI tools are designed to process and generate text based on existing data. The LLMs can write emails, summarize documents, generate code, and help with research. Yet, their intelligence is largely based on patterns, not true understanding.<\/p>\n

Add Zee News as a Preferred Source<\/p>\n


\n \"\"
\n <\/a><\/p>\n

Prof. Mishra explains that these systems do not \u201cthink\u201d or \u201ccreate\u201d in the human sense. Instead, they predict the most likely next word or idea based on massive datasets. This means they are excellent assistants\u2014but not replacements for human workers in most roles.
\n \n<\/p>\n

For example, an LLM can help draft a business proposal, but it cannot independently identify a completely new market opportunity or design a breakthrough business model from scratch.
\n \n<\/p>\n

Also Read:\u00a0The billion-dollar GPU boom may be over \u2014 Google just proved AI doesn\u2019t need all that hardware<\/a>
\n \n<\/p>\n

Why Most of Jobs Are Still Safe\u2014for Now
\n \n<\/p>\n

According to Mishra, most jobs involve a mix of routine and non-routine tasks. While LLMs can automate repetitive parts\u2014like documentation, data entry, or basic analysis\u2014they struggle with tasks that require judgment, creativity, or real-world context.
\n \n<\/p>\n

Look at professions like doctors, engineers, or journalists. AI can help them by speeding up research or drafting reports, but final decisions still depend on human expertise and accountability.
\n \n<\/p>\n

Even in creative fields, LLMs rely primally on existing content. They modify ideas rather than invent truly inventing original ones. This restrict their ability to disrupt jobs that rely on innovation.
\n \n<\/p>\n

\u00a0
\n \n<\/p>\n

\u00a0
\n \n<\/p>\n

<\/p>\n

Columbia CS Prof Vishal Misra explains why LLMs can\u2019t generate new science ideas.

\nBcz LLMs learn a structured map, Bayesian manifold of known data & work well within it, but fail outside it.

\nTrue discovery requires creating new maps, which LLMs can’t dopic.twitter.com\/PzI0YrTTeT<\/a>
\n\u2014 Rohan Paul (@rohanpaul_ai) April 21, 2026<\/a><\/p>\n

\u00a0
\n \n<\/p>\n

AGI: The Real Deal
\n \n<\/p>\n

The larger concern, Mishra notes, is Artificial General Intelligence (AGI)\u2014a future form of AI that could match or exceed human intelligence across a wide range of tasks.
\n \n<\/p>\n

Unlike LLMs, AGI would not just follow patterns. It would be capable of reasoning, learning independently, and generating genuinely new ideas. This is where the real disruption could begin.
\n \n<\/p>\n

For example, while an LLM can suggest improvements to an existing product, an AGI system could invent an entirely new product category\u2014something humans haven\u2019t even imagined yet.
\n \n<\/p>\n

Likewise, in scientific research, today\u2019s AI can help analyze data. But AGI could potentially form new theories, design experiments, and make discoveries without human guidance.
\n \n<\/p>\n

Also Read:\u00a0The more efficiently you use AI at work, the more work you will be given and that is not a coincidence<\/a>
\n \n<\/p>\n

Creativity: The Key Difference
\n \n<\/p>\n

One of the stark distinctions lies in creativity. LLMs are constrained by their training data. They can combine ideas in interesting ways but cannot truly \u201cbreak the mold.\u201d
\n \n<\/p>\n

AGI, on the other hand, is expected to go beyond this limitation. It could think conceptually, challenge assumptions, and generate original insights\u2014something that defines human intelligence today.
\n \n<\/p>\n

Mishra highlights that this ability to create new knowledge, rather than just process existing information, is what could eventually make AGI a serious competitor to human workers.
\n \n<\/p>\n

The Way Ahead
\n \n<\/p>\n

For now, the impact of AI is more about augmentation than replacement. LLMs are tools that enhance productivity rather than eliminate jobs entirely. Workers who learn to use these tools effectively may even gain an advantage.
\n \n<\/p>\n

However, the long-term picture could change dramatically if AGI becomes a reality. While that milestone is still uncertain, experts agree it would mark a fundamental shift in how work is done.
\n \n<\/p>\n

Until then, the message is clear: today\u2019s AI is a assistant, not a replacement. But tomorrow\u2019s AI could be something entirely different.
\n \n<\/p>\n

\u00a0<\/p>\n