Syrian journalists and Artificial Intelligence: “keeping abreast” despite restrictions

Muhammed Fansa | Hassan Ibrahim | Yamen Moghrabi | Hussam al-Mahmoud

Syrian media institutions are still far from the accelerating pace of the artificial intelligence revolution and are not effective in investing in it to develop various journalistic styles at a time when international media outlets are employing artificial intelligence in their work and in developing creative content and in front of the tendency of major technical companies to program applications that serve the traditional press.

The obstacles affecting AI in Syrian newsrooms are divided between logistical, technical, and knowledge obstacles, in addition to government restrictions that limit access to some of these tools for journalists residing inside Syria.

In this file, Enab Baladi discusses AI benefits for journalists and asks experts, academics, and journalists about its risks to the profession and the consequences of keeping pace with this development or not on media institutions.

Tools for journalists and organizations

Artificial intelligence has been associated with various fields, including the world of journalism, some of which benefit journalists in their daily work and some that benefit media corporations in providing better access to their audience.

Shehata al-Sayed, CEO of the Osh Group for Technology and Artificial Intelligence Industries, told Enab Baladi that the field of journalism and media is witnessing rapid developments with the advent of AI technologies and its use in this field and that by using appropriate tools, journalists can be helped to produce high-quality news content in a faster time.

One of the most important areas in which artificial intelligence tools benefit journalists is the ability to generate texts and preliminary content based on available data and facts, which saves time and effort in preparing reports. Also, summarizing tools allow journalists to quickly summarize long texts to extract key points.

Machine translation enables journalists to cover events from different regions in multiple languages accurately and quickly, while smart search tools provide the ability to search a huge amount of information to obtain relevant quotes and figures, according to al-Sayed.

Advanced tools allow verification of information and allegations, while other tools provide the ability to produce visual and audio content, including graphics, video clips, and audio, with high quality and lower cost.

Thanks to advanced forecasting and data analysis techniques, audience patterns and trends can be detected, which helps produce more in-depth and specialized press materials, according to al-Sayed.

Investigative journalist Abdul-Latif Haj Mohammad, a graduate of the London School of Economics in the field of discovering artificial intelligence and studying in the field of ethics in artificial intelligence, told Enab Baladi that in the midst of the “artificial intelligence revolution,” countless tools have emerged that can help journalists in different ways. Among them is the Bard tool that Google recently launched in Arabic, and it is one of the other tools the company is working on, according to Haj Mohammad.

Google is testing a tool known internally as Project Genesis that could take details of current events, for example, and generate news content from them, people familiar with the matter told The New York Times in July.

Google offered this tool to major American news organizations, such as The New York Times, The Washington Post, The Wall Street Journal, and News Corp, which owns several media outlets, which expressed concern about the “ridiculing” of this tool for the effort of journalists in producing accurate and creative material.

The AI tool, according to News Corp, is a kind of personal assistant for journalists to automate some tasks and save time while taking into account ethical responsibility, and that it will help guide the publishing industry away from errors of generative artificial intelligence such as ChatGPT.

Technologies towards better audience reach

The 2023 annual report of the Reuters Institute for the Study of Journalism highlighted the public’s tendency to receive information through social media, and with the development of technologies added to these platforms using artificial intelligence, the report recommended that the media focus more on determining how social media can be used to attract new people to follow.

The report warned of a wave of personalized news content driven by artificial intelligence but potentially “unreliable.”

AI technology can be used to create news summaries and personalized recommendations based on user preferences and behaviors, which can help journalists reach a larger audience and create more relevant and engaging news, according to investigative journalist Haj Mohammad.

Among the artificial intelligence systems currently available, Haj Mohammad sees three technologies that publishers should be looking for and media leaders should invest in:

1. Natural Language Processing (NLP): These are artificial intelligence systems that can read, write, translate, and generate texts, such as the ChatGPT platform that is based on Neuro-Linguistic Programming technology.

2. The recommendation system: It is an artificial intelligence technology based on machine learning and deep learning, and it can be used to customize sites. Examples of its uses include users receiving advertisements for products they searched for previously, and it is also used in article recommendations.

3. Computer vision or image recognition: It is a technology that has the ability to recognize images.

Through the latest technology, representatives from eight global news organizations worked on a project called “AIJO” (an abbreviation of the first two letters of artificial intelligence and the first two letters of journalism in English), which takes advantage of the power of artificial intelligence in several criteria, including, to understand, identify and mitigate newsroom biases, by identifying men and women in images, and calculating how equal they appear in reports and news, similarly to how much ethnic diversity is in the published news space.

In turn, the CEO of Osh Technology Group, Shehata al-Sayed, said that through predictive modeling technology, current news events and public opinion trends can be predicted using data, which allows the production of relevant content.

Search engines within the news site must be optimized to make them smarter to meet the search needs of readers, and bots can be used to automate routine tasks, such as publishing content and monitoring social media, according to al-Sayed.

Artificial intelligence (AI) refers to the simulation or approximation of human intelligence in machines. The goals of artificial intelligence include computer-enhanced learning, reasoning, and perception. AI is being used today across different industries, from finance to healthcare. There are currently many applications of artificial intelligence, the most important of which are expert systems, natural language processing, voice recognition, machine vision, or image recognition. Artificial intelligence as a term has been used in its current form since 2003 and is taught academically as a branch of computer science.

Barriers to the use of artificial intelligence in the Syrian press

Syrian media corporations and their journalists face many obstacles to using the new tools of artificial intelligence, which threatens the possibility of benefiting from them in developing their work and the content presented to the public, whether it is visual, audio, or written.

These obstacles are divided into three main sections: financial, technical, and training.

Arabic language, cognitive deficiency, and authoritarian grip

The technical barriers that organizations and independent journalists face are directly related to two main factors. The first is that most of the new tools do not support the Arabic language.

This may create a major obstacle for the journalist during his work, as well as the media not investing in departments where programmers, data analysts, and journalists meet to design and test tools that serve journalistic work, whether through artificial intelligence or otherwise.

According to the Syrian journalist Murad al-Quwatli, Syrian institutions, like their Arab counterparts, are waiting for and trying to use what is produced in the United States and Europe, but not all tools and programs that serve journalistic work reach the region, and if they do arrive, their use is restricted by the fact that they are tools of the institution that developed and produced them. Therefore, it is rare for an Arab media outlet to have its own artificial intelligence tool.

The CEO of Osh Group, Shehata al-Sayed, told Enab Baladi that the most prominent difficulties are the scarcity of AI tools and applications in the Arabic language compared to the English language in terms of morphology, parsing, and grammar and that the leading companies in this field are preoccupied with working in the English language, as it is a global and dominant language.

In addition to the foregoing, the poor accuracy of some machine translation tools and automatic transcription of Arabic texts due to the complexities of the latter, the lack of availability of data and the huge Arabic content needed to train artificial intelligence models, and the difficulty of accessing advanced algorithms, because a limited number of companies control them.

According to al-Sayed, these problems can be solved by encouraging Arab startups to develop artificial intelligence tools suitable for the Arabic language and the needs of media content, making huge data sets available to the private and public sectors, to train and develop artificial intelligence algorithms, and establishing joint laboratories between universities and media institutions, to research and develop the use of AI in journalistic work.

This requires providing the necessary technical infrastructure, including devices and networks, to accommodate these technologies, according to the AI expert.

Haj Mohammad sees continuous learning as an important tool in the hands of journalists, and they can allocate time to learn about what is new and changes that occur in the profession of journalism.

One of the obstacles is the process of training on artificial intelligence tools, which may require a certain experience to deal with them in order to reach the best possible results, and in light of the lack of access to all technologies to the Syrian and Arab press institutions, this seems difficult.

Al-Quwatli pointed out that Europe and the United States already have institutions that provide training on artificial intelligence tools, but the Middle East and North Africa region needs more time.

Al-Quwatli believes that media organizations and journalists in Syria and the Arab region are still new to learning about the capabilities of artificial intelligence in journalistic work, which began practically with the ChatGPT boom in 2022.

He added to Enab Baladi that this does not mean that there are many journalists who have come a long way in increasing their skills by using artificial intelligence tools and analyzing their strengths and weaknesses, and their experience can be used.

In addition to the above, there is no talk in the Syrian press or among Syrian journalists inside Syria about the use of artificial intelligence in their work, as the authorities block the websites of the most popular artificial intelligence tools, such as ChatGPT and Bard.

The Syrian authorities are constantly blocking various websites and pages on the Internet, most of which are opposition-affiliated websites, including some useful global websites such as social media, but they later lifted the ban on social media after popular pressure.

Financial barriers to independent institutions

Most AI tools offer paid services with annual or monthly subscriptions, and some of these tools require special training for journalists to learn to use them, which leads to additional financial burdens for organizations or freelance journalists.

These obstacles pose a challenge to independent press organizations, which may not receive significant annual funding from a specific entity on a continuous and permanent basis.

Al-Quwatli says that the high financial cost of artificial intelligence programs and tools poses a challenge to invest in them, explaining that artificial intelligence tools are not limited to ChatGPT, but there are programs and tools dedicated to analyzing data, audience behavior, etc., and the cost of subscription is high.

While press institutions in the region are facing great financial challenges, they also find themselves facing a new challenge represented in investing in artificial intelligence tools, whether by producing them or subscribing to paid services, according to al-Quwatli.

New AI tools play an active role in journalism and its development, which prompted al-Quwatli to create his own guide in which he collected these tools while analyzing their strengths and weaknesses, pointing out that journalistic institutions that do not pay attention to AI will find themselves very late in the next five years.

The CEO of Osh, al-Sayed, told Enab Baladi that the high cost of owning artificial intelligence tools and the difficulty of financing them by media institutions is one of the major difficulties faced by journalistic work.

The establishment of financial support funds and business incubators to support artificial intelligence projects is one of the solutions that institutions can follow to solve this problem, according to al-Sayed.

For his part, journalist Haj Mohammad said that only large media companies have the necessary financial resources to invest in expensive technologies and develop their own algorithms in this field.

While al-Quwatli believes that there are few solutions in this context, it remains for journalists and institutions to resort to using free copies as a temporary solution despite their limited usefulness and the inability to use them for free for a long time.

Journalists carrying out their daily duties in Enab Baladi’s newsroom – August 18, 2023 (Enab Baladi/Abdulmoeen Homs)

Does AI threaten journalists?

The speed of digital transformation threatened hundreds of thousands of jobs globally as the business world changes over time, and the future of work indicates an enlightened perspective on what companies and institutions need according to the data of digitization and technological developments amid the increasing revolution of robots and estimates of artificial intelligence solutions as a substitute for humans in many jobs.

According to a study by McKinsey and Company for global management consulting on the future of work for the year 2023, one in 16 workers may have to change their profession by 2030, representing more than 100 million workers across the economies of eight major countries, which together represent more than 60% of the world’s population.

The study considered that job growth will be more concentrated in jobs that require high skills, such as healthcare or science, technology, engineering, and mathematics.

In terms of journalism and media work, the Reuters Institute for Digital News 2023 report showed that the public prefers to get news from celebrities, influencers, and social media personalities rather than journalists and media professionals.

And the projections of the Center for Education and Workforce at Georgetown University in the United States showed that newsroom jobs, including visual and radio news anchor, reporter, and staff writer, will decrease by 3% from 2022 to 2031.

Auxiliary, not obsessive

Dr. Mona Abdel Maksoud, assistant professor at the Faculty of Mass Communication at Cairo University, told Enab Baladi that the future of journalistic work with artificial intelligence is a “promising” future due to its rapid development.

Abdel Maksoud believes that it saves the journalist from many tasks that used to consume his time and effort in order to produce routine work that does not require much creativity.

The assistant professor at the Faculty of Mass Communication gave an example that artificial intelligence will save the journalist the work of collecting and organizing information, dealing with big data, and revising it, as it was before, as the journalist currently relies on simultaneous translation, access to a number of information sources through open sources, as well as communication with decision makers and departments of officials through social media platforms.

In the part related to verifying information, Abdel Maksoud pointed out that the past period witnessed the development of a large number of tools that can help the journalist verify and revise the information and base it on evidence and data in order to start relying on it in monitoring.

Investing in AI in developing the journalist’s work will save a lot of effort and make him focus on creative and innovative matters, and any journalist who has not and will not develop his tools with or without artificial intelligence is writing the end of his journalistic career himself. Dr. Mona Abdel Maksoud – Assistant Professor at the Faculty of Mass Communication at Cairo University

Mattia Peretti, who manages JournalismAI — a research and training project at Polis, the international journalism think tank of the London School of Economics, considered that artificial intelligence does not steal journalistic jobs nor replace the journalist, but it can take on some tasks, and the user decides what they are.

Peretti stated that artificial intelligence can carry out several tasks that support the work of the journalist, and like any other technical innovation before, it changes the roles of the newsroom, but it is up to journalists after that to decide what they ask of artificial intelligence, as artificial intelligence does not have ambition in itself, nor the ability to steal jobs in the foreseeable future.

Dr. Arwa al-Kali, a trainer in data journalism, media, and news verification, believes that artificial intelligence cannot take over the work of information fact-checkers, but rather it can assist and guide them through various tools, and information-checkers are the ones who verify the authenticity of the news and the validity of its sources.

Al-Kali stated that it is not possible to rely on artificial intelligence to generate journalistic work because what it provides needs to be scrutinized, and it also provides false information and can confuse.

The trainer al-Kali’s assessment came during a training session that Enab Baladi attended virtually on information auditing and artificial intelligence, presented by the Pamela Howard Forum to cover global crises in cooperation with the International Journalists’ Network (IJNet).

After many information errors were discovered in AI platforms for generative conversations, the term “AI hallucinations” appeared, which describes chatbots fabricating the information provided, a problem that many technical experts doubt can be solved.

Keeping abreast of AI

The academic, Abdel Maksoud, expected that the journalist who is content with only collecting information manually, machine translation, or superficial media interviews, or even limiting himself to basic analytical skills and ordinary routine skills, will “disappear” and will have no place in the future of journalism.

Even with regard to critical thinking, problem-solving, innovation, and creativity, Abdel Maksoud suggested reaching a certain stage in the future in which artificial intelligence would be a competitor to humans, pointing to the need for people to be prepared for how to employ intelligence to serve their capabilities and capacities.

According to the Reuters Institute report on the trends and expectations of journalism, media, and technology for 2023, it is necessary to train journalists to work with artificial intelligence in order to avoid or reduce the effects of job displacement that it may cause.

The report pointed out the need for news organizations to invest in training journalists to work with artificial AI systems effectively and for news organizations, developers, and regulatory bodies to cooperate to set standards and guidelines for the ethical use of generative artificial intelligence in journalism.

Does it threaten journalists’ charter of ethics?

Various media outlets follow work mechanisms and professional charters that govern the manner in which they deal with various aspects of work.

Although these charters differ in their linguistic form, in line with the institution’s work style, identity, and the extent of its independence, at least six principles of journalism and media values constitute a common denominator that many media outlets seek to achieve in their coverage, namely accuracy, impartiality, objectivity, integrity, credibility, and independence.

In the face of any differences between what is narrow at the institutional level and what is broader at the professional level as a whole, in June 2019, the International Declaration of Professional Ethics for Journalists was adopted, which is an amended version of the Bordeaux Declaration issued in 1954, under the name of the International Federation of Journalists’ Declaration of Principles on the Conduct of Journalists.

The amended declaration consists of a preamble and 16 clauses specifying the rules and method for the journalist’s dealings with his sources and colleagues and the methods that can be used to obtain news and data, with the necessity for the journalist to acknowledge that fraud, distortion of facts, slander, spreading abuse and smearing reputation and unfounded accusations are a serious professional failure.

Articles 5 and 6 of the Declaration stipulated that the principle of speed in disseminating breaking information or news should not take precedence over the principles of verifying its sources and giving the concerned parties the opportunity to respond to them. The journalist should also make every effort to correct any information published that turns out to be inaccurate.

Faced with these principles and many others that fall within the same framework, workers in journalism and information verification stress the necessity of having rules for artificial intelligence that define what is available and possible and help the journalist in developing his tools and what is dangerous to the profession and affects its ethics in terms of undisciplined practices or uses.

The founder of the Verify-Sy platform, Ahmed Primo, told Enab Baladi that talking about standards governing the uses of artificial intelligence is not premature, as misleading, inaccurate, or fake content using artificial intelligence techniques is present and widespread, and major media organizations and journalists have fallen into the trap, and the content has been treated as accurate content, like the pictures that spread of the arrest of former US President Donald Trump.

In Primo’s opinion, these matters added a burden on fact-checkers and journalists to raise awareness for users of communication sites, media institutions, and journalists.

In terms of artificial intelligence means to combat misleading content, there are some techniques that people may use for research motives that do not start with curiosity and do not stop at research purposes, which provide exaggerated content, completely subjective, what calls for concern about the future of journalism or information, whether as journalists or contemporaries of a new experience present strongly.

It is necessary to develop media codes of honor and standards that journalists follow with regard to AI, and this takes a long time to reach our Arab region and a longer time for the journalist and the recipients of information to understand it at the same time. Ahmed Primo – Founder of the Verify-Sy platform

The recipient is exposed daily to a lot of misinformation with the emergence of content that is not subject to the standards of accuracy and objectivity.

However, access to artificial intelligence will facilitate the spread of the phenomenon of “fake journalists,” “fake programmers,” and “fake artists.” These tools do not mimic everyone’s professions, but they will turn many into incompetent people in their professions, according to Primo.

The issue requires a lot of effort, awareness, and caution, as the image today can lie, and the sound can be imitated, or an exact copy can be created, according to the founder of the Verify-Sy platform.

Primo stressed the need to develop tools for fact-checkers and those affected by this phenomenon to come up with means that enable them to discover content based on this technology and distinguish it from unreal content, with the need to spread more media awareness and create curricula related to media education to teach it, similar to curricula for erasing technology illiteracy.

Investigative journalist Abdul-Latif Haj Mohammad agrees with Primo that the development of some programs to a large extent in generating and organizing information does not negate the role of the journalist in verifying the information he receives, without denying that there are basics in verifying information, photos, videos and different types of data that some journalists have not realized yet.

Haj Mohammad also indicated the need to work on a comprehensive guide that outlines the broad lines for the use of artificial intelligence, as artificial intelligence systems can have a built-in bias by creators that leads to prioritizing certain stories or viewpoints over others or neglecting human groups on political grounds, for example, and stifling its views.

Journalist Murad al-Quwatli believes that holding discussions at the level of press institutions is a means of determining the mechanisms for using artificial intelligence and identifying the ethical and professional aspects in dealing with this technology.

This necessitates cooperation between journalistic institutions, specifically in a region such as the Middle East, in a way that serves to advance journalistic work and agree on the basis for new tools in the profession, concludes al-Quwatli.

Enab Baladi asked AI

ChatGPT is one of the most popular artificial intelligence tools, which is used in many fields and sectors, including journalism.

Enab Baladi asked ChatGPT what it can offer to journalists, and its answer was as follows:

1- Idea generation: The tool can help journalists exchange ideas about articles, headlines, and news story angles by providing a brief general idea on a specific topic and requesting a model for generating creative ideas.

2- Research assistance: The tool can be used to collect basic information or statistics and contexts for reports and articles by requesting the provision of data, facts, or explanations related to the requested topic.

3- Drafting and planning: It can help draft outlines or initial drafts of articles by asking for expansion in different sections or obtaining supporting details.

4- Preparing for press interviews: The tool can be used to simulate interview questions and answers. This helps anticipate possible responses and improve the final result.

5- Editing and Proofreading: The tool helps review and edit written content, suggest sentences with alternative wording, and correct grammatical and spelling errors.

6- Accuracy: The tool helps verify the validity and accuracy of information before it is presented to the public.

7- Explain complex topics: The tool can explain complex topics to journalists in simple ways, which is reflected in the materials presented to the public later.

8- Breaking news: ChatGPT can generate quick summaries based on the available information, which helps in the completion of breaking news.

9- Interactive content: The tool helps create interactive elements such as surveys and tests.

10- Translation: It provides fast translation from several different languages.

ChatGPT concluded that although it is a valuable tool for journalists, it is essential to verify the information received and for the journalist to be aware of ethical considerations and potential biases within the information.