Well, what’s going on with the Linkedin song?

One day in November, the pharmacist called Michelle (not her real name), her drill), to her Linkedin account and exchanged men. He also changed his name to Michael, he told Techcrnn.

He was working on a project called #weateatants right there Women were tested Until the new Linkedithm Algorithm has started to catch on with women.

For many months, some rich Users have complained that they are seeing dots dating to cartoons on the internet. This came after the company’s president, At Jerda, said in August that the tower “Soon” written “which was established LLMs to support a useful environment for users.

Michelle (who goes by the name Techcrnchnch) was skeptical about the change because she has a strong following with her husband, who is only 2000 years old. However, she and her husband like to hang out with the same criminals, she said, even though he was very old.

“The most important thing was women,” he said.

Marilynn Jousner, the founder, also changed her profile or gender. She has been posting on Linkedin “I changed my profile on my profile from female to male, and my views jumped 238% in no time,” she told Techcrunch.

Megan so.

Linecin he said “Algorithms and AI Systems do not use information about age, race, or gender to determine the nature of what is not fair, or whether it is” discrimination to feed or imply that there is no injustice.

Algorithm experts agree that mild sexism is not a factor, although discrimination is at work.

The tower is “a complex piece of algorithms that draws on mathematics and the nature of mathematics, simultaneously and frequently.” Brandois Marshalldata consultant, said the technology.

“The change of the image and the name of a person is the only one,” he said, adding that the algorithm is also influenced, for example, by how the user lives and currently interacts with other things.

“What we don’t know are the other similarities that make this algorithm think so much about one person. This is a much more difficult problem than people think,” he says.

Brod

A #Yalipants Testing began with two experts – gallop and Jane Evans.

They asked two men to make and send the same thing as them, curious to know if most men and women have the problem. Gallop is both enuws they have ugly consequences – More than 150,000 compared to two men who had about 9,400 at the time.

Gallop says his post will reach only 801 people, While the father who wrote The best content reached 10,408 people, more than 100% of his followers. Then other women took part. Others, such as Ashe, who uses LinkedIn to market his business, were affected.

“I’d like to see Linkedin take no credit for any fun that might be inside its algorithm,” Joyner said.

But LinkedIn, like other LLM search and networking platforms, provides detailed information about content and related content.

Marshall said that many of these platforms would “fill in a white, thuggish, Western-Centered image” because of those who teach the models. The researchers found Evidence of Human Imagery as Sexuality and Choice Popular LLM types because types are taught by humans, and People often participate in teaching or learning to learn.

However, the way each company implements its AI system is based on a secret algorithmic black box.

LinkedIn says the #yapatharants test did not show up between genders vs. women. Words of August he continued to another post In November – that his system is not using information as a sign of appearance.

In fact, LinkedIn told tetsynch that it tests millions of people in the afternoon to connect with opportunities. It said Demographics were only used to test this, such as seeing posts “from creative component makers and what you see in food,” the company told Techcrnch.

LinkedIn has been identified Search and Change its algorithm attempt to provide a limited contract user experience.

It’s an unknown variable, Marshall said, which may explain why some women have seen it increase their male profile. For example, being exposed to a virus, can lead to a person being involved in an activity; Some accounts were posting for a long time, and the algorithm could reward them for that.

The tone and tone of the writing can also play a role. “Michael,” Michelle said, changing her tone slightly, writing directly, as the man did. That’s when he said:

He concluded the plan was not sexist, “but it appeared that it was a communication agency often associated with women” to create a low price. ”

Stereotypical men Writing styles are believed to be concisewhere Writing Spriotypes for women it is thought to be simple and general. If the LLM is trained to be able to refer to what is related to the male mind, which is a subtle unconscious. I am As we said beforeresearchers have determined that most LLMs are filled with them.

Sarah Dean, an assistant professor of computer science at Cornell, said that platforms like LinkedIn often use the text, including what makes it possible. This includes activity on the user’s profile and the type of activity they engage in.

“Someone’s people can affect ‘both sides of the algorithm – what they see and what they see,'” Dean said.

LinkedIn told unccrunch that AI Its AI machine looks at signals Antings to determine what is forced on the user, including the recognition of a person’s profile, network, and work.

“We’re constantly trying to understand what helps people get what’s most important, at the right time for their jobs,” he said. “Member behavior also changes the feed, what people click, save, and change or how they like.”

Chad Johnson, a sales professional working at LinkedIn, explainers Modifiers such as mocks, comments, and trolls. The LLM program “doesn’t care how often you write or what time of day it is,” Johnson wrote in the time of day. “It matters if your writing shows understanding, clarity, and relevance.”

All of this makes it difficult to determine what happens as a result of #yakiti.

People just don’t like it much

However, it seems like many people, in most age groups, either don’t like or don’t understand the new algorithm – whatever it is.

Shailvi Wakhuzu, a data scientist, told zatekinonch that he produced one word per day for five years and used thousands of hacks. Now she and her husband have the opportunity to see several hundred. He said: “Weary because of contentment is very honest,” he said.

One man told Techcrnchn he saw about 50% going a few months ago. However, one man said that he sees him show and reach more than 100% in the same period. “This is because I write on certain topics, which is what this new algorithm is beneficial for,” he told Techcrnnnch, adding that his customers are seeing this increase.

But in Marshall’s case, he, who is black, believes the records of his experiences do better than the institutions. “If black women only talk about black women but not when they talk about their art, then that’s bias,” she said.

The researcher, Dean, believes the algorithm can only improve “everything that already exists.” It can be beneficial in some cases, not because of the author’s profile, but because there is already a history of response on the platform. While Marshall may have been disappointed in some beautiful area, his anecdotal evidence does not determine this for sure.

LinkedIn has figured out what works best now. The company said that users have grown, and thanks to the service, sending is up 15% last year while reviews are up 24% yoy. “This means more competition in food,” the company said. Articles related to health professionals and education, career news and analysis, and education or information to do around work, business, and finance are all doing well, it said.

If anything, people are confused. “I like to impress,” says Michelle.

However, being satisfied things are secrets always protect secrets in their companies, and their respect can lead to seconding them, that is a big question. It is one that has never been enough.