Meta’s New AI Models Aim to Tackle Inaccuracies in Technology
Meta Platforms (META) has introduced several new artificial intelligence (AI) models, including a groundbreaking design that audits other AI outputs for accuracy.
The model, named “Self-Taught Evaluator,” is part of a set of releases from Meta announced on October 18. It is able to assess responses produced by chatbots and various AI applications in challenging fields like science, coding, and mathematics.
Meta’s goal is to address a recurring criticism aimed at AI models such as ChatGPT: the tendency to deliver incorrect or outdated information. The company claims to have developed methods that simplify complex problems, leading to enhanced AI responses.
Paving the Way for Self-Improving AI
Experts highlight that using AI to evaluate other AI represents a significant advancement toward creating autonomous AI systems. This evolution suggests models could eventually learn from their own errors, self-correct, and enhance their functionality without human guidance.
Although the concept of self-improving technology may sound like something out of a science fiction narrative, analysts believe it marks progress toward achieving “artificial general intelligence.” This is a theoretical point where AI would match human intelligence and make independent decisions. Some researchers predict we could reach this stage as early as 2030.
On October 18, Meta also unveiled updates to its image-identification software and introduced a tool designed to expedite AI response times. Notably, META stock has surged by 64% this year.
Analyzing META Stock Potential
Meta Platforms stock has received a consensus Strong Buy rating from 47 Wall Street analysts. This rating stems from 42 Buy recommendations, four Hold calls, and one Sell rating issued over the past three months. The average price target for META is $624.12, suggesting a potential upside of 7.91% from current prices.
Read more analyst ratings on META stock
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
