AI, Translation And The Holy Grail Of "Natural Language"

Important digital innovations have been put into practice in the areas of translation, subtitling and text-to-image.

AI, Translation And The Holy Grail Of "Natural Language"

The Google Translate logo displayed on a smartphone —

David Larousserie and Alexandre Piquard

PARIS — When asked about advances in language management through artificial intelligence, Douglas Eck suggests pressing the "subtitle" button on Meet, the video conferencing service used for the interview, because of the COVID-19 pandemic. The words of this American engineer, who had come to Paris to work at Google's French headquarters, were then displayed in writing, live and without error, under the window where we see him, headset on. This innovation, unthinkable until recently, is also available for most videos om YouTube, the Google subsidiary. Or on the dictaphone of its latest phones, which offers to automatically transcribe all audio recordings.

These new possibilities are just one example of the progress made in recent years in natural language processing by digital companies, especially giants such as Google, Apple, Facebook and Amazon (GAFA). Some of these innovations are already being put into practice. Others are in the research stage, showcased at annual developer conferences, such as Google I/O (which took place May 18-20) and Facebook F8 (June 2).

Artificial intelligence generates 20 billion translations per day on Facebook

In the crucial area of translation, services such as Google Translate, which has expanded its offer to 104 languages, or the German competitor DeepL now make it possible to translate entire paragraphs in a coherent and fluid manner. Thanks to these advances, Google offers to translate the subtitles of YouTube videos.

Facebook has also come a long way. Artificial intelligence generates 20 billion translations per day on the social network (several dozen languages, including Wolof, are available), compared to only six billion in 2019.

Yann LeCun, Facebook's chief artificial intelligence scientist and a pioneer in the field, says, "This area is very important for Facebook. And we know that simultaneous translations in real time will be possible."

The dream of a machine translating live conversations is within reach. Google Translate comes close, but with a slight delay: You can speak in a language and have the other person hear or read the translation via a smartphone, and even listen to their translated response through headphones, if they are the latest in-house models.

The barriers between text and image are disappearing. With the augmented reality application Google Lens, students can scan a page from a textbook or a handwritten sentence with their smartphone and translate it or get additional information online. A tourist can understand a sign or a menu or get information about a monument.

It's all because software has learned to recognize subjects in images. Tomorrow, we could launch a search with a photo, Google believes. The American company OpenAI is exploring the creation of images from a text description. Its DALL-E prototype offers disturbing representations of invented objects: an alarm clock in the shape of a peach, a pig lamp...

These innovations help make digital technology more accessible to the disabled and illiterate. With the French National Institute for Research in Digital Science and Technology (Inria), Facebook is studying the simplification of forms, with pictograms and synonyms. In January, the company presented an automatic image description tool for the blind and visually impaired. Google has a voice recognition project for people with speech difficulties, called "Euphonia."

Now, artificial intelligence is capturing more complex sentences than before. Amazon claims that its voice assistant Alexa has, by 2020, learned to understand more variations around simple dialogue, ask questions about unknown words and even anticipate a user's "intent" — and suggest a timer, if they ask for tea brewing time. The Google search engine answers questions like: "Where does the Seine begin?" or "What political party is the newspaper Libération from?"

Pandu Nayak, vice president of search at Google, says, "We could, in the long run, handle queries with complex intentions." For example: "I've already climbed Mount Adams, and I want to climb Mount Fuji, how do I prepare?" The answer would be broken down into sub-queries, with links to training tutorials, gear, maps, videos or content translated from Japanese, although this work remains "very conceptual."

This wave of innovation is enabled by recent scientific breakthroughs in machine learning or deep learning. This technology competes with humans in the game of Go or image recognition. Its principle is to adapt the billions of parameters of a program in order to propose the best association between a set of known "questions" and "answers." In 2017, Google invented a new way to organize them to improve machine translations. Called "Transformer," it was quickly adopted by Facebook, the Chinese Baidu, the French-American Systran and the German DeepL.

"The last time there was such a breakthrough was five years ago, with long short-term memory [LSTM] architectures, used in voice assistants," says Douglas Eck, the Google engineer. In 2018 "self-supervised learning" was added to this breakthrough: Google showed that a Transformer could do without human supervision to "learn" a language. Until now, however, to find the right value for the program's parameters, software needed vast databases annotated by humans.

The system, named "BERT" (Bidirectional Encoder Representations from Transformers), "learns" to fill in blank sentences, then proves to be excellent in grammar exercises, questions and answers. It inspired Facebook's system, named "Roberta;" then OpenAI's GPT-3, with its 15 billion parameters and 500 billion words ingested (100 times the English version of Wikipedia); and the Chinese system Wu Dao 2.0, which is already 10 times larger.

Thomas Wolf, co-founder of Hugging Face, a company specializing in the distribution of these models, says "Faced with too much information and text, we will increasingly need these language models to find our way around."

The prospects are promising, but also dizzying because these technologies will be used in headphones, in homes, in cars. The concerns have been gathered in an article co-authored by Timnit Gebru and Margaret Mitchell, two researchers in ethics whose dismissal by Google has caused controversy. The main concern is about the "biases" — racist, sexist, homophobic — that these softwares can reproduce, or even amplify, after training on masses of texts from the internet.

For example, chatbots can slide toward conspiracy themes. In response, Google says it excludes "offensive" parts of the web, such as certain forums, from its training data. "As these systems grow, it's our responsibility to make sure they stay fair," says Eck. Although he prefers to look for solutions "on a case-by-case basis" depending on usage, rather than trying to correct all biases in the datasets. Others, including the international collective Big Science, want to use a better documented and less biased body of text.

The main concern is about the "biases" — racist, sexist, homophobic — that these softwares can reproduce.

Another limitation of such software is its focus on the most used languages on the internet, which makes it less effective on "low-resource" languages, due to a lack of training data. Digital giants are trying to mitigate this imbalance. In October 2020, Facebook presented a software capable of translating 100 languages, without going through English, which is currently mandatory.

The "big language models" are also denounced for their gigantism. Gebru and Mitchell point out an environmental risk related to the energy consumption involved, even if the figures are debated. The authors, who describe current software as "stochastic parrots," say that investments should be made in less data-intensive models.

All agree that these systems do not really "understand" the language. "The results are sometimes bluffing, but we also see that the generated texts end up containing errors that are easy to see. These systems have no common sense or knowledge of the world, unlike children," says Yann LeCun, Facebook's chief AI scientist, who is looking for ways to improve.

In the meantime, the impressive growth of computer-assisted language will be accompanied by increasing questions. Discussions around social issues — will children continue to learn foreign languages — will be joined by regulatory debates. Like any complex algorithm, these software programs will have to be made more transparent and understandable. We can anticipate questions of responsibility in case of error: A Palestinian man was arrested because of a mistranslation in a Facebook post, Gebru points out.

The automatic moderation of content by algorithms can also infringe on freedom of expression and is the subject of regulation projects. In its rules on artificial intelligence adopted in April, the European Commission proposes to adapt the framework according to the level of risk involved. For example, it recommends that internet users be informed when they are conversing online with a software program and not with... a human.

Keep up with the world. Break out of the bubble.
Sign up to our expressly international daily newsletter!

How Thailand's Lèse-Majesté Law Is Used To Stifle All Protest

Once meant to protect the royal family, the century-old law has become a tool for the military-led government in Bangkok to stamp out all dissent. A new report outlines the abuses.

Pro-Democracy protest at The Criminal Court in Bangkok, Thailand

Laura Valentina Cortés Sierra

"We need to reform the institution of the monarchy in Thailand. It is the root of the problem." Those words, from Thai student activist Juthatip Sirikan, are a clear expression of the growing youth-led movement that is challenging the legitimacy of the government and demanding deep political changes in the Southeast Asian nation. Yet those very same words could also send Sirikan to jail.

Thailand's Criminal Code 'Lèse-Majesté' Article 112 imposes jail terms for defaming, insulting, or threatening the monarchy, with sentences of three to 15 years. This law has been present in Thai politics since 1908, though applied sparingly, only when direct verbal or written attacks against members of the royal family.

But after the May 2014 military coup d'état, Thailand experienced the first wave of lèse-majesté arrests, prosecutions, and detentions of at least 127 individuals arrested in a much wider interpretation of the law.

The recent report 'Second Wave: The Return of Lèse-Majesté in Thailand', documents how the Thai government has "used and abused Article 112 of the Criminal Code to target pro-democracy activists and protesters in relation to their online political expression and participation in peaceful pro-democracy demonstrations."

Criticism of any 'royal project'

The investigation shows 124 individuals, including at least eight minors, have been charged with lèse-majesté between November 2020 and August 2021. Nineteen of them served jail time. The new wave of charges is cited as a response to the rising pro-democracy protests across Thailand over the past year.

Juthatip Sirikan explains that the law is now being applied in such a broad way that people are not allowed to question government budgets and expenditure if they have any relationship with the royal family, which stifles criticism of the most basic government decision-making since there are an estimated 5,000 ongoing "royal" projects. "Article 112 of lèse-majesté could be the key (factor) in Thailand's political problems" the young activist argues.

In 2020 the Move Forward opposition party questioned royal spending paid by government departments, including nearly 3 billion baht (89,874,174 USD) from the Defense Ministry and Thai police for royal security, and 7 billion baht budgeted for royal development projects, as well as 38 planes and helicopters for the monarchy. Previously, on June 16, 2018, it was revealed that Thailand's Crown Property Bureau transferred its entire portfolio to the new King Maha Vajiralongkorn.

photo of graffiti of 112 crossed out on sidewalk

Protestors In Bangkok Call For Political Prisoner Release

Peerapon Boonyakiat/SOPA Images via ZUMA Wire

Freedom of speech at stake

"Article 112 shuts down all freedom of speech in this country", says Sirikan. "Even the political parties fear to touch the subject, so it blocks most things. This country cannot move anywhere if we still have this law."

The student activist herself was charged with lèse-majesté in September 2020, after simply citing a list of public documents that refer to royal family expenditure. Sirikan comes from a family that has faced the consequences of decades of political repression. Her grandfather, Tiang Sirikhan was a journalist and politician who openly protested against Thailand's involvement in World War II. He was accused of being a Communist and abducted in 1952. According to Sirikhan's family, he was killed by the state.

The new report was conducted by The International Federation for Human Rights (FIDH), Thai Lawyer for Human Rights (TLHR), and Internet Law Reform Dialogue (iLaw). It accuses Thai authorities of an increasingly broad interpretation of Article 112, to the point of "absurdity," including charges against people for criticizing the government's COVID-19 vaccine management, wearing crop tops, insulting the previous monarch, or quoting a United Nations statement about Article 112.

Juthatip Sirikan speaks in front of democracy monument.

Shift to social media

While in the past the Article was only used against people who spoke about the royals, it's now being used as an alibi for more general political repression — which has also spurred more open campaigning to abolish it. Sirikan recounts recent cases of police charging people for spreading paint near the picture of the king during a protest, or even just for having a picture of the king as phone wallpaper.

The more than a century-old law is now largely playing out online, where much of today's protest takes place in Thailand. Sirikan says people are willing to go further on social media to expose information such as how the king intervenes in politics and the monarchy's accumulation of wealth, information the mainstream media rarely reports on them.

Not surprisingly, however, social media is heavily monitored and the military is involved in Intelligence operations and cyber attacks against human rights defenders and critics of any kind. In October 2020, Twitter took down 926 accounts, linked to the army and the government, which promoted themselves and attacked political opposition, and this June, Google removed two Maps with pictures, names, and addresses, of more than 400 people who were accused of insulting the Thai monarchy. "They are trying to control the internet as well," Sirikan says. "They are trying to censor every content that they find a threat".

Keep up with the world. Break out of the bubble.
Sign up to our expressly international daily newsletter!