Hello and welcome to the eye for artificial intelligence. In this edition … the media is struggling with artificial intelligence; Trump orders us to make safety efforts from artificial intelligence to re -focus on combating “ideological bias”; Distributed training acquires a traction. Amnesty International can increase increasingly strong standards towards totalitarianism.
Artificial intelligence is likely to be turbulent to the business models of many organizations. In a few sectors, however, it is apparently the existential threat like news works. It happens that this is the work that I am, so I hope a somewhat newsletter will forgive. But the news must be important to all of us because the operating company free Press plays a fundamental role in democracy – with the public’s performance and assistance in calculating power. There are some similarities between how the executives express the news – decisively, from the challenges and opportunities offered by Amnesty International from which business leaders in other sectors can learn as well.
Last week, I spent one day at the Asbin Institute conference entitled “AI & News: Drawing the session”, which was hosted at Reuters headquarters in London. The conference was attended by senior executives from a number of news organizations in the United Kingdom and European. It was held according to the rules of Chatham’s house, so I cannot tell you exactly what he said, but I can transfer what was said.
Tools for journalists and editors
News executive officials spoke about the use of artificial intelligence in the first place in internal confrontation products to make their teams more efficient. Artificial intelligence helps in writing improved engine titles and translating content-allowing institutions greatly Reaching new fans In places that were not traditionally served, although most of them stressed the keeping of humans in the ring to monitor accuracy.
One of the editors using artificial intelligence to produce short articles automatically described journalists, and editing journalists for more original reports, while maintaining the human editors to monitor quality. Journalists also use Amnesty International to summarize documents and analyze large data collections – such as government documents dumps and satellite images – delivery of investigative journalism that will be difficult without these tools. These are good use, but they lead to a modest effect – most of them about making current workflow tasks more efficient.
From bottom to top or from top to bottom?
There was an active discussion between the leaders of the news rooms and the technicians on whether the news institutions should follow an approach from the base to the top-to filter the tools “Code of atmosphere” Self -powered efforts to help them in their jobs should be, or whether efforts are from top to bottom, with management giving project priorities.
The approach from bottom to top contains advantages-weakening access to democratic intelligence, and the frontal lines who often know pain points, often can discover good use cases before high-level executives can talent artificial intelligence developer to be spent only on projects that are larger, more complicated, and strategically more important.
The negative aspect of the approach from bottom to top is that it can be chaotic, which makes it difficult for the organization to ensure compliance with moral and legal policies. It can create artistic debts, as tools are built on flying that cannot be easily preserved or updated. One of the editors is concerned about creating a two -level dual news room, as some editors embrace new technology, and others are backward. The base not also guarantees the summit that solutions generate the best return on investment-where artificial intelligence models can become expensive. Many called for a balanced approach, although there is no consensus on how to achieve this. From the conversations I had with Execs in other sectors, this dilemma is familiar across the industries.
A warning of danger
Newswear is also cautious about building artificial intelligence tools facing the audience. Many have begun to use artificial intelligence to produce lead points summaries for articles that could help readers and are increasingly patient. Some have built chat groups that can answer questions about a certain narrow sub -group of coverage – like stories About the Olympic Games Or climate change– But they were tending to naming these “experiments” to help refer to readers that the answers may not be always accurate. Few went further in terms of the content created from artificial intelligence. They are concerned that Gen Ai will undermine confidence in the accuracy of their journalist. Their brands and companies ultimately depend on that confidence.
Those who hesitate will they get lost?
This caution, although it is concept, is in itself a tremendous danger. If the news institutions themselves do not use artificial intelligence to summarize the news and make it more interactive, then technological companies are. People are increasingly turning into search engines in artificial intelligence and chat chat, including confusion, Openai’s Chatgpt, Gews’s Gemini and “AI Overviefs” that Google is now in response to many searches, and many of them. Many news executives at the conference said that “instability” – the loss of direct relationship with their fans – was their biggest fear.
They have anxiety. Many news organizations (including luck) At least it depends on Google’s research to bring the masses. A recent study by Tollbit– Who sells programs that help protect web sites from web crawling – Google AI clicks were 91 % less than the traditional Google search. (Google has not used an artificial intelligence overview of news inquiries, although many believe it is only a matter of time.) Other studies to click through the rates of Chatbot conversations are equal. CloudflareIt is also provided to help protect news publishers from web scraping, I found it Openai has shipped a 250 -time news website for each display of one referral page that you sent to this site.
To date, news organizations have responded to this potential existential threat through a combination of legal recovery – New York Times It was filed a lawsuit against Openai for copyright violations, while Dow Jones New York Post A lawsuit against confusion– And partnerships. These partnerships included multi -numbers lying deals for news content. ((luck He has a partnership with both confusion and prorta.) Many executives at the conference said that licensing deals were a way to achieve revenues from the content, most likely “technology companies” have already stolen anyway. They also saw partnerships as a way to build relationships with technology companies and benefit from their experiences to help them build AI products or train their employees. Nothing saw relationships as especially stable. They were all aware of the risk of becoming more than that on the revenue of artificial intelligence licensing, after they were previously burned when the Facebook media industry allows it to become a major engine of traffic and advertising revenues. Later, this money disappeared in practice overnight Dead After 2016, CEO Mark Zuckerberg decided to cancel the confirmation of news in people’s summaries.
Ferrari made an intelligent time with the same
Executive managers acknowledged the need to build direct relationships from the public that cannot be compensated by artificial intelligence companies, but a few of them have clear strategies to do so. “The news industry does not take artificial intelligence very seriously”, with a focus on “gradual adaptation rather than structural transformation.” The current methods are likened to a three -step process that contains “Acting Ferrari” on both sides, but “a horse and a vehicle in the middle”.
He and another media consultant in the media industry urged news organizations to stay away from the structure of their news in the news about “Articles”. Instead, they encourage the executives to think about the ways in which materials can be converted (general data, interview texts, documents obtained from the sources, raw video clips, audio recordings, and archival news stories) into a variety of directors-live broadcast, short video, lead points summaries, or yes, a traditional news article-to fit the audience’s tastes to unify technology. They also urged news organizations to stop thinking about producing news as a written process, and start thinking about it as a circular link, and there was no human being in the middle.
One person at the conference said that news organizations need to become less isolated and closely look at visions and lessons from other industries and how they were adapting to artificial intelligence. Others said that it may require emerging companies – perhaps the news organizations themselves – to manufacture new business models for the era of artificial intelligence.
The risks cannot be higher. Although artificial intelligence introduces existential challenges to traditional journalism, it also provides unprecedented opportunities to expand access and reconnect to the audience who “stop the news” – if the leaders are bold enough to re -imagine what the news could be in the era of artificial intelligence.
However, here is more news of artificial intelligence.
Jeremy was
[email protected]
Jeremyakahn
revision: Tuesday edition last week from Eye on AI He made a mistake in identifying the country in which TrustPilot is located. It is Denmark. Also, the news component in that edition has determined the name of the Chinese youth behind the viral artificial intelligence model. The starting name is the butterfly effect.
This story was originally shown on Fortune.com
https://fortune.com/img-assets/wp-content/uploads/2025/03/GettyImages-2147904415-e1742324771791.jpg?resize=1200,600
Source link