China Internet Court Attributes AI Generated Image Copyright To Human Prompt Creator

On Monday, the Beijing Internet Court held that a human plaintiff prompt is sufficient to invoke copyright protection in a Stable Diffusion generated image, so long as the output qualifies as an “original” work. Copyright is determined on a case by case basis, so this decision is not entirely inconsistent with other AI jurisprudence trends favouring human input, as we saw in the US Thaler v. Perlmutter decision denying copyright registration for a work created absent any human involvement. In the Beijing case, there was sufficient human involvement in the form of words and a specific combination thereof thought up by the plaintiff which became the the basis of enforceable copyright in artificially generated images.

Last February, in Zarya of the Dawn (Registration # VAu001480196) in response to a registration request for a comic book made with Midjourney generated images, the U.S. Copyright Office also recognized that the registrant is the author of the Work’s text as well as the selection, coordination, and arrangement of the Work’s written and visual elements, however fell short of recognizing copyright in the comic book’s images, because the current registration for the Work did not disclose its Midjourney-generated content”, so a new certificate of registration was issued that covers only the expressive material created by the registrant. Having looked at the comic book, I believe that the new certificate indeed covers all the copyright there is to cover in this comic book. The same comic book could use an entirely different set of AI generated images and would still qualify as the same work. It is also possible at the face of the Copyright Office Decision that, had the registrant disclosed the use of AI generated software, the certificate would have covered those images as well. So from my perspective, there are no flagrant inconsistencies in the US and China cases involving AI generated works.

Beijing Case Background

The plaintiff, Mr. Li, used Stable Diffusion to generate the image involved in the case by inputting prompt words and published it on the Xiaohongshu platform under the name “Spring Breeze Brings Tenderness”. A week later, the defendant, a blogger on Baijiahao, removed plaintiff’s signature watermark from plaintiff’s AI generated image and used it to accompany their article “Love in March, in the Peach Blossom”. The plaintiff sued claiming that the defendant did not obtain the plaintiff’s permission and given the removed signature causing users to mistakenly believe that the defendant was the author of the work, which seriously violated the plaintiff’s right of attribution and information network dissemination rights. 

Analysis

The Court held that the artificial intelligence-generated image involved in the case met the requirements of “originality” and reflected a human’s original intellectual investment and should be recognized as works and protected by copyright law. The Court stated that from the time the plaintiff conceived the image involved in the case to the final selection of the image involved, the plaintiff made a certain amount of intellectual investment, such as designing the presentation of characters, selecting prompt words, arranging the order of prompt words, and setting relevant parameters and so on which reflected the plaintiff’s choice and arrangement on a continuum of time. It didn’t happen from a first try. The images involved in the case reflected the plaintiff’s important intellectual investment, aesthetic choices and personal judgments, and in that regard the images involved in the case met the requirements of “intellectual achievements.”

The pictures involved were not just “mechanical intellectual achievements.” In the absence of contrary evidence, the court concluded that the images involved in the case were independently completed by the plaintiff and reflected the plaintiff’s personalized expression, for the images involved in the case to meet the requirements of “originality.”

Conclusion

The Court awarded 500 RMB ($96CAD) to the plaintiff in damages and costs of 50 RMB.

The full decision courtesy of 知产库 is available here (Chinese only).

Fuite de données personnelles d\’anciens employé(e)s d\’Olymel

Olymel affirme que les renseignements personnels d’employés actuels, d’anciens employés et de postulants ont pu être dérobés.

\”Les données concernées sont liées au dossier d’emploi et comprennent le nom, l’adresse, la date de naissance, le numéro d’assurance sociale et, dans certains cas, pour les travailleurs étrangers, des informations contenues dans leurs passeports.\”

https://ici.radio-canada.ca/nouvelle/1769537/cyberattaque-olymel-loi-renseignements-personnels

3.2B Email and Password Pairs Leaked

CyberNews reports that a massive repository of individuals’ data has been posted online. Dubbed “Compilation of Many Breaches” (COMB) this may be the biggest-ever compilation of hacked user credentials ever posted online before, but it’s not the result of a new hack or data breach. Leaked user data combines a 2012 data breach at LinkedIn involving 117 million accounts, as well as stolen Netflix login data. This is the time to change passwords and it comes as a reminder for users to stop recycling user names and passwords across different services.

https://bgr.com/tech/data-breach-email-and-passwords-leaked-compilation-of-many-breaches-5904287/

Le télétravail exige une réforme ciblée

Selon le Syndicat de professionnelles et professionnels du gouvernement du Québec (SPGQ), le projet de loi 59 censé moderniser le régime de santé et sécurité du travail souffre d\’une déficience majeure: il fait abstraction du télétravail.

https://www.tvanouvelles.ca/2021/01/26/projet-de-loi-59-un-syndicat-deplore-labsence-de-balises-encadrant-le-teletravail

CPRA Into CPAA, California Privacy Upgraded

The CPRA introduces amendments to the CCPA of existing provisions of Title 1.81.5 of the California Civil Code (currently known as the CCPA and codified at Cal. Civ. Code § 1798.100 et seq) and adds new provisions (related to the establishment California Privacy Protection Agency). It is unclear, however, whether Title 1.81.5 will continue to be known as the CCPA or will instead be known as CPRA effective Jan. 1, 2023.

The CPRA took effect on Dec. 16, 2020, but most of the provisions revising the CCPA won’t become “operative” until Jan. 1, 2023.

Here is a diagram breaking down the two statutes and outlining how the CPRA expands the CPAA.

\"\"

[For additional information, see Bloomberg Law\’s Glossary of Terms for Decoding CCPA/CPRA.]

The California Privacy Protection Agency is a new agency, created by the CPRA, which is vested with “full administrative power, authority, and jurisdiction to implement and enforce” the CCPA. The CPRA transfers rulemaking authority from the California Attorney General to the California Privacy Protection Agency effective July 1, 2021, with final CPRA regulations due by July 1, 2022.

https://news.bloomberglaw.com/bloomberg-law-analysis/analysis-california-privacy-reboot-puts-rights-in-spotlight

WhatsApp Controversy

In response to Apple’s new data disclosure requirements, WhatsApp informed users last week that certain data points, such as the user’s profile status, login activity, contact list, purchases, and financial information, may be shared with businesses and the third parties they use.

Unless you consent to totally unacceptable 3rd party data sharing of your personal and financial information, your account will be deleted anyway.

Therefore delete it before it deletes you.

https://www.fastcompany.com/90593066/whatsapp-facebook-privacy-ultimatum This article discusses one of the most unhinged antitrust defendants in the history of litigation.

The alternative to Whasapp is Telegram or Signal. The simple reason that big tech anti-trust defendants are backing Signal at the moment however should come as ared flag. Until all anti-trust lawsuits against Twitter, Apple, Amazon, Google, and FB are resolved, it is impossible to trust any company so heavily endorsed by social media tyrants.

Even Turkey\’s Erdogan has dropped Whatsapp urging journalists to use a Turkish app equivalent.


I maintain my position that government actors have absolutely no business to be on private social media networks. If you are a politician on Twitter, you are giving off an image of selling out your country to Big Tech.

It would be reckless to encourage your citizens to hand over their personal information to big tech giants for the privilege to follow you on private social media networks.

From a political standpoint, private companies have 100% the right and the power to pick and choose sides and ban whoever they want. Users are being tolerated as guests on these platforms and must behave by their code. If your opinion falls out of line, they can legally remove you.

Governments need to set up their own communication platforms. Otherwise, they call into question their legitimacy and raison d\’être. Governments that keep relying on private corporations to communicate hand government power to non elected private actors. It is like building your house on someone else\’s land. The landowner can kick you out at any time. As a government actor, you need a more stable alternative.


Twitter: The Great Detox. Keep it UP!

I am welcoming Twitter\’s initiative to suspend government and lawyers accounts. Ideally all political content should be taken down from the platform. Over the past 5 years, every politician on this planet has been imitating the US president who started this tweeting circus as the chief commander of most tweets per second.

This should\’ve happened long ago. The US president\’s phone should\’ve been confiscated the minute he was sworn in.

Every single politician, rather than working for the people, has been chasing after the user engagement leveraged by the US president who, being the most hated person in America, naturally has the most followers, beginning with mainstream media who are hooked on the president. They\’re already experiencing withdrawal symptoms. It shows.

Media outlets whose reporting has been almost entirely about the president\’s tweets now have nothing to show for 5 years of reporting!

Tweeting lawyers are another curious bunch. One thing you learn in litigation is to abstain from making predictions on evidence or outcomes, because even in the clearest of cases, stuff happens. Announcing what will happen is such a beginner\’s mistake.

Take them all down! All politicians should be banned from social media. ALL.


CBC is already embedding Parler declarations (or whatever you call them) authored by one of the president\’s lawyers who was banned from twitter. Withdrawal symptoms are kicking in… Imagine when cold turkey hits Parler too.

Lastly, have you noticed how representing certain clients never ends well for the lawyer! The idea that everyone is entitled to representation must be revisited. It is chilling how representing this president has annihilated so many lawyers. Some have landed in prison, forced to breach lawyer client confidentiality in exchange for a plea deal, others sued in defamation for billions of dollars, all systemically hit with disbarment complaints…

Is there a professional rule of conduct that warns against taking up a \”cursed client\”? We need one fast.

The idea of putting aside your own interest to fearlessly raise every legal argument, no matter how distasteful, so as to zealously represent a client appears to be the exact conduct that may send you to a disbarment hearing these days.


Update January 10, 2020. The other side of the coin

Nothing stops politicians to post on their own websites or create their individual platforms owned and controlled by government and politicians themselves. So long as they set up their own servers, they\’re good to go.

When it comes to the president, he was the actual \”talent\” on twitter. The app itself needs people like him to be profitable, it was losing money before allowing politicians in. If the president sets up his own platform, his followers (especially haters) will follow him wherever he goes. For obvious reasons, private corporations should never again be allowed to host government and state related information.

I welcome total decentralization of information

Since yesterday, I have discovered at least 50 websites I hadn\’t heard of before and counting. Nothing stops you from following the people who are permanently banned from twitter, simply by bookmarking their websites and set alerts.

Your search history belongs to you. Law enforcement authorities need a warrant to use your search history against you. Social media companies on the other hand have deals with law enforcement that are entirely out of your control. Your precious personal and engagement data on twitter and facebook is being traded and shared with countless other third parties without your knowledge or consent. The logic behind it is that you already consented simply by signing on the platforms, by accepting the terms and conditions. Nobody reads terms and conditions, therefore nobody really knows what they consent to.

To avoid unintended uses of your personal data, I strongly recommend to abstain from using social media apps on your phone. For advanced protection inspired from the country of the rising sun (Japan) is to give platforms fake information about yourself, not your real name. Create an online persona through an alternative email (not your main email) with completely different identifiers from your legal identifiers. For example set your birthdate to 1901 and choose a different gender. This is how it is done in Japan, you never mix your real persona with your online personas. It is the only way that your real persona won\’t suffer the consequences of fame caused by your online personas.

The people of China are many great examples on how to communicate through coded messages and entirely circumvent censorship. I have learned so much.

At the end of the day, there will always be dark forums operating on military browsers like Tor. I strongly recommend exploring those. I also anticipate other military communication methods to become mainstream through 2021.

Ontario Human Rights Code On Psychometric Tests At Hiring


En Ontario, administrer des tests psychologiques avant l\’embauche est considéré une violation de la vie privée et un motif valable de plainte pour discrimination. Le fait de ne pas être dotés d\’une codification équivalente au Québec ne veut pas dire que ces arguments ne se plaident pas ici. L\’Ontario est d\’ores et déjà une juridiction aux lois très étoffés et qui donne toujours plus d\’information et de ressources à ses citoyens en matière d\’accès à la justice.

http://www.ohrc.on.ca/en/iv-human-rights-issues-all-stages-employment/6-requesting-job-related-sensitive-information

Quelques surlignés:

Au tout début de la page, on conseille aux employeurs de porter une attention particulière à la liste qui énumère les documents de nature confidentielle (dont les tests psychologiques), et de s\’assurer d\’avoir fait une offre d\’emploi valable avant d\’administrer de tels tests.

Sous la rubrique c) on discute de ces tests, mais je trouve intéressante aussi la partie b) pour des précautions relatives aux tests médicaux.

En ce qui concerne les tests psychométriques:

\”Tests should be tailored to actual job duties.\” = doivent être adaptés (faits sur mesure) pour les tâches de l\’emploi

\”Avoid testing that seeks to assess personal interests, attitudes and values. If these tests are legitimately needed to assess ability to perform a job, use them with great care to make sure they do not favour certain cultures or genders. Many such tests are outdated and may have been created based on stereotypes or biases relating to Code grounds.\”

il faut éviter d\’évaluer les intérêts personnels, attitudes et (surprise) VALEURS personnelles. Plusieurs de ces tests datent d\’un certain temps qui les rend incompatibles avec les sensibilités d\’aujourd\’hui, ou ils sont carrément bourrés de stéréotypes qui peuvent constituer des motifs de discrimination suivant le Human Rights Code (équivalent de la Charte des droits et libertés)

\”Even if a test is fair, an employer will need to put in place measures to minimize the impact of unintentional bias on the part of persons scoring candidates’ answers. One option is to have more than one person score each candidate.\”

même si le test n\’est pas discriminatoire (selon la bonne foi de l\’employeur), l\’employeur doit mettre en place des mesures de sécurité afin de minimiser l\’impact de préjugés non-intentionnels de la part des personnes qui évaluent le candidat (les 3rd parties)

Subsection 23(2) of the Code prohibits the use of an employment application form or a written or oral inquiry that directly or indirectly classifies an applicant on the basis of a prohibited ground of discrimination. This also applies to psychological profiles and testing. The validity of behavioural testing as a tool to predict on-the-job performance may be subject to a complaint under the Code.