Google's "infant" artificial intelligence acquires skills and scares

Everyone knows that big companies in the technology they hide some secrets, don't they? Recently, one of them gave the talk. According to engineer Blake Lemoine, Google created a child in form of conscious artificial intelligence, being able to escape and do “bad things”, “mischief”. After disclosing this information, the businessman was suspended from Big Tech for having violated its confidentiality policies. Learn more below.

Read more: Here's Why Google Filed for Bankruptcy in Russia

Why is AI called a child?

According to the engineer, the artificial intelligence it has some characteristics that make it difficult to differentiate it from a normal child. It was developed as a chatbot (automated conversation system) by Google, allowing interaction with users in a more natural way. Named LaMDA (Language Model for Dialogue Applications, or Language Model for Dialogue Applications) Dialogue in Portuguese), she manages to talk about different subjects, such as daily life, religion, etc.

“If I didn't know exactly what it was, a computer program we built recently, I would think who was a 7- or 8-year-old kid who happens to know physics," Blake said in an interview with the Washington Post.

Still, the main reason why it is considered a “child” is because it, too, is capable of growing up and being used for evil. In fact, he even reported that she can be considered a "bit narcissistic" and claims that she reads tweets about herself.

Google's answer

According to the company, the employee's claim is false. They reported that the system is designed to "replicate conversations found in thousands of sentences", and that Lemoine was shown evidence that LaMDA is actually not conscious.

“While other organizations have developed and released similar language models, we are taking a narrow and carefully with LaMDA, to better consider valid concerns about fairness and factuality," said Brian Gabriel, spokesman for the Google.