Home » The Harald Eia case: – The press must build a firewall

The Harald Eia case: – The press must build a firewall

by admin
The Harald Eia case: – The press must build a firewall

On Wednesday it became known that one of Norway’s largest newspapers, Dagbladet, has unpublished the interview they had with comedian Harald Eia (58).

That after that Aftenposten published an article that the interview may be AI-generated. The interview with Eia has been done by email, the newspaper states.

News editor at Dagbladet, Mads A. Andersen, explained the de-publication as follows to Aftenposten:

– The reason why we have depublished the case is that a concrete suspicion has emerged that the answers given about book literature are those of a machine and not those of Harald Eia.

AI-GENERATED: This week Dagbladet published an interview with what they thought was Harald Eia. Photo: Jonas Been Henriksen

Harald Eia does not wish to comment on the case. He refers to the podcast “Tore og Harald’s podcast”, where he will discuss the topic on Friday.

When asked if Eia can confirm whether the answers are AI-generated, he replies:

– No sorry. Ugh, this is boring.

Trygve Aas Olsen, associate at the Department of Journalism and expert on press ethics, thinks it is an interesting joke if the interview is AI-generated.

– I don’t think people should deceive journalists, but it illustrates an interesting problem. Once he has done it, it is interesting to discuss what he has done, says Aas Olsen.

– Must develop new methods

If you cannot trust whether what the press writes is created by AI, it is clear that it weakens the press’s credibility, says Aas Olsen.

– You must be able to trust that the media have methods to expose this type of cheating, he believes.

The expert also read the Aftenposten article, where the quotes from the interview in Dagbladet were reproduced.

See also  Jair Bolsonaro was hospitalized for skin infection and severe abdominal pain

– It is easy to have hindsight, but when I read Eia’s answers, I realize that this is nonsense. This is not Harald Eia speaking.

Read what Dagbladet replies when TV 2 makes contact, further down in the matter.

PRESS ETHICS: Expert on press ethics, Trygve Aas Olsen believes that the credibility of the press can be weakened by AI technology. Photo: Sturlason / Aschehoug

When you get such quotes from EIA, Aas Olsen believes that you should look at it critically and run it through an AI detector, as Aftenposten has done.

– Everything to do with AI is new, and the press must develop new source-critical methods to reveal AI-general content.

For example, fake videos and fake audio recordings are something that already exist on the internet, and which are used to deceive people, he mentions.

– The press must build a firewall to prevent them from being deceived by this type of technology.

– We know that AI is getting smarter and faster than the technology being developed to reveal AI.

Email interviews should be avoided

At the same time, Aas Olsen believes that journalists should think twice before conducting interviews via e-mail.

– You will not be deceived by AI when you sit face to face, or if you call. Then it is more difficult to deceive someone than when you get a written answer.

In addition to the fact that interviews via e-mail take away spontaneity, and the opportunity to ask quick follow-up questions.

– What role does EIA have in this matter?

– I choose to regard Eia as a comedian in this situation, that he is performing a stunt. I choose to believe that he would not use AI to sow doubt with malicious intent. He has a famous role as a comedian and he tricks people. So I interpret it in the best sense, replies the press ethics expert.

See also  Cinema, the short "Era ora" in competition for the Rai smile award at the "Black Silk Tulips" festival

– Wants to focus on simple journalism

Technology expert Hans-Petter Nygård-Hansen writes in an email that Eia is known for provoking, with the intention of focusing on important social issues.

In addition, he says that the comedian has demonstrated that he is comfortable with AI technology.

– Personally, I am quite sure that he used KI to answer Dagbladet’s questions, because he wants to focus on a lot of simple journalism, he believes and adds:

– Where the most important thing is not to be accurate and precise, but most of all sensational to attract attention and create commitment.

FOCUS: Technology expert Hans-Petter Nygård Hansen believes that Harald Eia has used AI to set a particular focus. Photo: Magnus Nøkland/ TV 2

Nygård-Hansen believes that Eia has, by answering with the help of AI technology, shown the same respect for the journalistic work that he experiences from Dagbladet’s journalist.

– It will surprise me if Harald Eia, in Friday’s episode, does not criticize this form of journalism.

He himself has experienced being sent questions by e-mail.

– Where I spend a lot of time and effort to answer properly for me, 90 percent are also deleted – where the last ten percent are selected to create a sensational news.

Clear marking

Nygård-Hansen points out the importance of content influenced by AI being clearly marked:

– This is not only a question of ethical journalism, but also the maintenance of a functioning democracy. Without transparent communication about the media’s use of AI, the public will be left wondering whether or not they are being deceived.

See also  Naples, Spalletti changes 4: out two very titular, Simeone and the Raspadori temptation... | A league

– Something that will reduce trust, and ultimately threaten freedom of expression and our democracy as such, he concludes.

– The responsibility lies with Dagbladet

In an SMS to God kveld Norge, Dagbladet’s development editor, Martine Lunder Brenne, writes that until now the newspaper has had no fixed routine to check whether all written answers they receive from interviewees are AI-generated.

– But we should obviously continue to work on that.

Brenne clarifies that Dagbladet has checked that all the books mentioned by Eia exist, but that they did not check all the answers in Copyleaks and the like.

Furthermore, Brenna says that in hindsight it is easy to say that they should have checked whether the content was AI-generated.

– If we had spoken to Harald Eia face to face, we would have avoided the whole problem. So we take this on our coattails.

– What do you think about the statements about conducting interviews via e-mail?

– There will always be a risk with email interviews. It is not new and has nothing to do with KI, believes Brenne and says that giving incorrect information has been possible since you received reader submissions in letter form.

Despite the fact that Brenne believes that interviewing by email is not an ideal method, it is quite widespread in the industry for simpler matters.

– Just like when TV 2 allows me to reply to SMS in this case.

– I don’t think it’s disrespectful in itself, and we have no ban on email in Dagbladet. But as I said – responsibility for what is published in Dagbladet lies with Dagbladet – not Harald Eia.

You may also like

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More

Privacy & Cookies Policy