Protect Your Data with Runbox: Secure, Private, and Based in Norway

We live in an era of constant digital surveillance, where governments and corporations collect vast amounts of personal data. Privacy has become one of the most pressing issues for people around the world. From targeted ads to government surveillance programs, personal information is constantly at risk. Protecting that privacy is not just a matter of convenience — it’s essential to safeguarding our freedoms, security, and autonomy. Runbox’s base in Norway plays a pivotal role in safeguarding your personal information.

(more…)

Continue Reading →

EU’s AI Act: What it is all about

From the perspective of the general public and society as a whole, so-called Artificial Intelligence (AI) was largely invisible until OpenAI removed the veil over GPT-3.5/ChatGPT in 2022.

Since then, the interest and use of AI, and General Purpose AI (GPAI), has exploded. AI implementations are creeping in everywhere, to great benefit in many respects. However, the warning signs are many — manipulated images, fabricated conversations, falsified news stories, and fake video-presented events can lead to unforeseeable negative consequences, for instance in influencing democratic elections.

This is also the case when AI is used to make decisions, since we know that there is always a risk of “AI hallucinations” where AI software produces incorrect or misleading information.

This simplified outline can serve as background for the EU’s AI Act, whose purpose is to put a societal control over the use and influence of AI/GPAI in particular, and the big tech companies in general.

So let’s dig into the matter, with the aim to describe what EU AI Act is, and to clarify its consequences, if any, for Runbox.

(more…)

Continue Reading →

The Grindr case illustrates how Norwegian authorities fight against misuse of personal information

Oslo District Court has found Grindr’s sharing of personal data illegal as a result of the Norwegian Consumer Council complaint from 2020. Accordingly, Grindr has to pay EUR 5 million, as fined by the Council.

Our guardians of personal data and privacy: NDPA, NPAB, and NCC

As we have written multiple times in our blog series about GDPR and consequences of this EU-regulation, Norway has a long history of protecting citizens’ personal information. It started out with the first Personal Data Act implemented in 1978 with the purpose of protecting the individual against privacy being violated through the processing of personal data. The law was updated with GDPR clauses in the year 2000.

In 1980, the Norwegian Data Protection Authority (NDPA) was established as an independent authority whose task is to monitor compliance with the Personal Data Act. It is important to note that the NDPA has two roles: supervisory authority and ombudsman.

The NDPA decisions may be appealed to NPAB, Norwegian Privacy Appeals Board (Personvern­nemda), whose decisions are final.

During recent years, another Norwegian governmental public body, the Norwegian Consumer Council (NCC), whose role is to protect consumers’ interests, has become involved in privacy, more precisely the misuse of personal data that big tech companies are involved in. As a governmental-independent agency, the NCC is free to chose the cases they want to work on.

Sharing of personal data is illegal without specific consent: The Grindr case

Recently, the NCC has put effort into the task of preventing the big tech companies from using personal information for surveillance-based marketing that the users have not consented to. Neither have users given consent to how personal data is transmitted to the companies’ partners.

The figure below, from https://noyb.eu/en/eu-58-million-fine-grindr-confirmed, illustrates the problem.

(more…)

Continue Reading →

Be privacy concerned when using ChatGPT (and other AI chatbots)

This is blog post #18 in our series on the GDPR.

“Don’t tell anything to a chatbot you want to keep private.” [source]

Writing about AI in general and about chatbots specifically is like shooting at a moving target because of the speed of development. However, at Runbox we are always concerned about privacy and must examine the chatbots case in that respect.

Due to its popularity, we have mainly used ChatGPT from OpenAI as the target of our examination. NOTE: ChatGPT and the images from text captions DALL-E are both consumer services from OpenAI.

This blog post is a summary of our findings, leading to advice on how to avoid putting your privacy at risk when using the Natural Language Processing (NLP)-based ChatGPT.

Our examination is based on OpenAIs Privacy Policy, Terms of Use, and FAQ, and a number of documents resulting from hours of Internet browsing.

The blog post consists of two parts: PART I is a summary of our understanding of the technology behind language models in order to grasp the concepts and better understand its implications regarding privacy. In PART II we mainly discuss the relevant privacy issues. It is written as a stand alone piece, and can be read without necessarily have read PART I.

PART I: Generative AI technology

The basics

GPT stands for Generative Pre-trained Transformer, and GPT-3 is a 175 billion parameter language model that can compose fluent original writings in response to a short text prompted by a user. The current version of ChatGPT is built upon GPT-3.5 and GPT-4 from OpenAI.

ChatGPT was launched publicly on November 30, 2022. ChatGPT was released as a freely available research preview, but due to its popularity, OpenAI now operates the service on a freemium model [source].

The GPTs are the result of three main steps: 1) Development and use of the underpinning technology Large Language Models (LLMs), 2) Collection of a very large amount of data/information, and 3) Training of the model.

Let us also keep in mind that all this is possible only because of today’s advancements of computational power.

Language models

A language model is a system which denotes mathematics “converted” to computer programs that predict the next word/words in a sentence, or a complete sentence, based on probabilities. The model is a mathematical representation of the principle that words in a sentence depend of the words that precede them.

Since computers basically can only process numbers (in fact only additions and comparisons), text input to the model (prompts) must be converted to numbers, and likewise the output numbers have to be converted to text (response). Text in this context consists of phrases, single words, or parts of words called tokens.

When prompting a GPT then, your query is converted to tokens (represented by numbers), and used by the transformer where its attention mechanism generates a score matrix that determines how much weight should be put on each word in the input (prompt). This is used to produce the answer to the prompt, using the model’s generative capability – that is to predict the next word in a sentence by selecting relevant information from the pre-processed text with high level of probability of being fluent and similar to human-like text [source].

The learning part of the model is handled by a huge number of parameters representing the weights and also statistical biases for preventing unwanted associations between words. For instance, GPT-3 has 175 billion parameters, and GPT-4 is approximated to have around 1 trillion.

(The label “large” in LLM refers to the number of values (parameters) the model can change autonomously as it learns.)

Collecting the data

The texts the GPT model generate stems from OpenAIs scraping of some 500 billion words (in the case of GPT-3, the predecessor for the current version of ChatGPT) systematically from the Internet: books, articles, websites, blogs – all open and available information, from libraries to social media – without any restriction regarding content, copyrights or privacy.

The scraping includes pictures and program codes as well and is filtered resulting in a subset where “bad” websites are excluded

The pre-training process

All that data is fundamental for pre-training the model. This process analyses the huge volume of data (the corpus) for linguistics patterns, vocabulary, grammatic properties etc. in order to assign probabilities to combinations of tokens and combinations of words. The aforementioned transformer architecture is used in the training process, where the attention mechanism makes it possible to capture the dependencies between words independent of their position in a sentence.

The result of the pre-training process is an intermediate stage that has to be fine-tuned to the specific task the model is intended for, for instance providing texts, program code, or translation of speech as response to a prompt. The fine-tuning process uses appropriate task-specific datasets containing examples typically for the task in question, and the weights and parameters are adjusted accordingly.

Of cause, a ChatGPT-response to a prompt is not “burdened” with the ethical, contextual, or other considerations a human will perform. To prevent undesired responses (toxicity, bias, or incorrect information), the fine-tuning process is supervised by humans in order to correct inappropriate or erroneous responses, using prompt-based learning. Here the responses are given a “toxicity” score that incorporates human feedback information [source].

ChatGPT usage training

The learning process continues when response generated following by a user’s prompts is saved and subject to the training process, at least for 30 days, but “forever” if chat history isn’t turned off. In any event it is not possible to delete specific prompts from user history [source], only entire conversations

In the world of AI and LLMs, hallucinations are the word used when responses are like “pulled from thin air”.

OpenAI offers an API that makes it possible for “anyone” to train GPT-n models for domain specific tasks [source], that is to build a customized chatbot. In addition, they have launched a feature that allow GPT-n to “remember” information that otherwise will have to be repeated [source, source].

Takeaways

  • The huge volume of data scraped is obviously a cacophony of contents and qualities that will affect the corpus and so also the probability pattern and the responses produced [source].
  • ChatGPT has limited knowledge of events that have occurred after September 2021, the cutoff date for the data it was trained on [source].
  • The response you get from ChatGPT to your prompt is based on probabilities, and as such you have no guarantee of the validity [source].
  • A prompt starts a conversation, unlike a search engine like DuckDuckGo and Google that gives you a list of websites matching your search query [source].
  • ChatGPT uses information scraped from all over the Internet, without any restrictions regarding content, copyrights, or privacy. However, manual training of a model was introduced to detect harmful content [source]. Violations of copyrights has resulted in lawsuits [source], and also signing of more than 10 000 authors of an open letter to the CEOs of prominent AI companies [source].
  • Your conversation is normally used to train the models that power ChatGPT, unless you specifically opt-out [source].

PART II: Chatbot privacy considerations

The privacy considerations with something like ChatGPT cannot be overstated” [source]

The following introduction is mainly made for readers that have skipped this blog post PART I.

Generative AI systems, such as ChatGPT, use information scraped from all over the Internet, without permissions nor restrictions regarding content, copyrights, or privacy (more on this in PART II). This means that what you have written on social media, blogs, comments on an article online etc. may have been stored and used by AI companies to train their chatbots.

Another source for training of generative AI systems is prompts, that is information from users when asking the chatbot something. What you ask ChatGPT, the sentences you write, and the generated text as well, is “taken care of” by the system and could be available for other users through the answer of their questions/prompts.

However, according to Open AI’s help center article, you can opt-out of training the model, but “opt-in” is obviously default.

So, both the Internet scraping and any personal information included in your prompts can have as result that personal information could turn up in a generated answer to another arbitrary prompt.

This is very problematic for several reasons.

Is Open AI breaching the GDPR?

First, OpenAI (and other scraping of the Internet) never asked for permission to use the collected data, which could contain information that may be used to identify individuals, their location, and all kinds of sensitive information from hundreds of millions of Internet users.

Even if Internet scraping is not prohibited by law, it is ethically problematic because data can be used outside the context in which it was produced, and so can breach contextual integrity, which has de facto been manifested in the EU’s General Data Protection Regulation (GDPR) Article 6, 1 (a) as prerequisite for lawful processing of personal data:

…the data subject has given consent to the processing of his or her personal data for one or more specific purposes

Here language models, like Open AI’s ChatGPT, are in trouble: Personal data can be used for any purpose – a clear violation of Article 6.

Second, there is no procedures given by Open AI for individuals to check if their personal data is stored and thereby can potentially be revealed by arbitrary prompt, and far less can data be deleted by request. This “right to erasure” is set forth in the GDPR Article 17, 1:

The data subject shall have the right to obtain from the controller the erasure of personal data concerning him or her without undue delay …” on the grounds that “(d) the personal data have been unlawfully processed

It is inherent in language models that data can be processed in ways that are not predictable and presented/stored anywhere, and therefore the “right to be forgotten” is unobtainable.

Third, and without going into details, the GDPR gives the data subjects (individuals) regarding personal data the right to be informed, the right of access, the right to rectification, the right to object, and the right to data portability. It is questionable if generative AI systems can ever accommodate such requirements since an individual’s personal data could be replicated arbitrarily in the system’s huge dataset.

Fourth, Open AI stores all their data, including personal data they collect, one way or another, on servers located in the US. That mean they are subject to the EU-US Data Privacy Framework (see our blog Privacy, GDPR, and Google Analytics – Revisited), and the requirements set there.

To answer the question posed in the headline of this paragraph, Is OpenAI breaching the GDPR?It is very difficult to understand how ChatGPT, and other language models for generative use (Generative AI systems) as well, can ever comply with the GDPR.

What about the privacy regulations in the US?

Contrary to the situation in Europe, there is no federal privacy law in the United States – each state has their own jurisdiction in this area. There are only federal laws such as HIPAA (Health Insurance Portability and Accountability Act) and COPPA (Children’s Online Privacy Protection Act) which regulate the collection and use of personal data categorized as sensitive. However, there are movements towards regulation of personal information in several states as tracked by IAPP (The International Association of Privacy Professionals).

How do OpenAI use data they collect?

When signing up to ChatGPT, you have to agree to OpenAI’s Privacy Policy (PP), and allow them to gather and store a lot of information about you and your browsing habits. Of course, you have to submit all the usual account information, and to allow them to collect your IP-address, browser type, and browser settings.

But you also allow them to automatically collect information about for instance

“… the types of content that you view or engage with, the features you use and the actions you take, as well as your time zone, country, the dates and times of access, user agent and version, type of computer or mobile device, and your computer connection”.

All this data made it possible to build a profile of each user – bare facts, but also more tangible information such as interests, social belongingness, concerns etc. This is similar to what search engines do, but ChatGPT is not a search engine — it is a “conversational” engine and as such is able to “learn” more about you depending on what you submit in a prompt, that is, how you engage with the system. According to their PP and the citation above, that information is collected.

The PP acknowledges that users have certain rights regarding their personal information, with indirect reference to the GDPR, for instance the right to rectification. However, they add:

“Given the technical complexity of how our models work, we may not be able to correct the inaccuracy in every instance.”

OpenAI reserves the right to provide personal information to third parties, generally without notice to the user, so your personal information could be spread to actors in OpenAI’s economic infrastructure and is very difficult to control.

Misuse of your personal information – what are the risks?

It is reasonable to assume that OpenAI will not knowingly and willfully set out to abuse your personal information because they have to adhere to strict regulations such as GDPR, where misuse could result in fines of hundreds of millions of dollars.

The biggest uncertainty is linked to how the system responds to input in combination with the system’s “learning” abilities.

If asked the “right” question, the system can expose personal information, and may combine information about a person, e.g. a person’s name, with characteristics and histories that are untrue, and which may be very unfortunate for that individual. For instance, asking the system something about a person by name, can result in an answer that “transforms” a credit card fraud investigator to be a person adhered to credit card scam.

Takeaways

Using generative AI systems, for example ChatGPT, is like chatting with a “black box” – you never know how the “box” utilizes your input. Likewise, you will never know the sources of the information you get in return. Also, you will never know if the information is correct. You may also receive information about other individuals that you shouldn’t have, potentially even sensitive and confidential information.

Similarly, other individuals chatting with the “box”, may learn about you, your friends, your company etc. The only way to avoid that, is to be very careful when writing your prompts.

That said, OpenAI has introduced some control features in their ChatGPT where you can disable your chat histories through the account settings – however the data is deleted first after 30 days, which means that your data can be used for training ChatGPT in the meantime.

You can object to the processing of your personal data by OpenAI’s models by filling out and submitting the User Content Opt Out Request form or OpenAI Personal Data Removal Request form, if your privacy is covered by the GDPR. However, when they say that they reserve the right “to determine the correct balance of interests, rights, and freedoms and what is in the public interest”, it is an indication of their reluctance to accept your request. The article in Wired is recommended in this regard.

Valuable sources

  1. GPT-3 Overview. History and main concepts (The Hitchhiker’s Guide to GPT3)
  2. GPT-3 technical overview
  3. Transformers – step by step explanation
  4. LMM training and fine-tuning

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.

Continue Reading →

On the EU-US data transfer problem

This is blog post #16 in our series on the GDPR.

At Runbox we are always concerned about data privacy – “privacy is priceless” – and we put some effort into keeping ourselves updated on how the EU’s General Data Protection Regulation (GDPR) affects privacy related issues.

That’s because we want to be prepared in case something happens within the area that will affect the Runbox organization, our services, and consequently and most important: our customers.

Update 2023-08-06

On 10 July the European Commission adopted its adequacy decision for the EU-US Data Privacy Framework and announced a new data transfer pact with the United States. See the full text here: COMMISSION IMPLEMENTING DECISION.

A flyer from the European Commission is also available, and a summary of the situation is available from Reuters.

The Austrian non-profit organization NOYB, chaired by Maxmillian Schrems, stated:

“We now had ‘Harbors’, ‘Umbrellas’, ‘Shields’ and ‘Frameworks’ – but no substantial change in US surveillance law. The press statements of today are almost a literal copy of the ones from the past 23 years. Just announcing that something is ‘new’, ‘robust’ or ‘effective’ does not cut it before the Court of Justice. We would need changes in US surveillance law to make this work – and we simply don’t have it.“

“We have various options for a challenge already in the drawer, although we are sick and tired of this legal ping-pong. We currently expect this to be back at the Court of Justice by the beginning of next year. The Court of Justice could then even suspend the new deal while it is reviewing the substance of it.” [source]

The last words are obviously not said.

Originally published 2023-03-19

The case of EU-US data transfer is highly relevant because Runbox has an organizational virtual modus operandi, and that this could lead to an opportunity to involve consultants that are residing in the US. We know that many of our customers are as concerned as we are about data privacy, so we believe it is pertinent to share our findings.

In blogpost #15 in our series of the GDPR we referred to the Executive Order signed by US President Joe Biden on 07 October 2022. This happened six months after the US President and the President of the EU Commission Ursula von der Leyen with much publicity signed the Trans-Atlantic Data Privacy Framework on 25 March 2022.

Joe Biden and Ursula von der Leyen at a press conference in Brussels. [Xavier Lejeune/European Commission]

In this blog post we will take a closer look at the Trans-Atlantic Data Privacy Framework, and the process thereafter.

Trans-Atlantic Data Privacy Framework

The objective of the Framework is to (re)establish a legal (with regards to the GDPR) mechanism for transfers of EU personal data to the United States, after two former attempts (Safe Harbour and Privacy Shield) were deemed illegal by the Court of Justice of the European Union (CJEU).

The Framework ascertains United States’ commitment to implement new safeguards to ensure that ‘signals intelligence activities’ (SIGINT, intelligence-gathering by interception of signals) are necessary and proportionate in the pursuit of defined national security objectives. In addition, the Framework commits the US to create a new mechanism for EU individuals to seek redress if they believe they are unlawfully targeted by signals intelligence activities.

Following up the 25 March 2022 Biden–von der Leyen agreement, the US president signed on the 7 October 2022 the Executive Order (EO) ‘Enhancing Safeguards for United States Signals Intelligence Activities’.

US compliance with the GDPR

Subsequently a process was initiated on 13 December 2022 within the EU Commission to assess whether the US, after the implementation of the EO, will meet the requirements qualifying the US to the list of nations that is compliant with the GDPR Article 45 “Transfers on the basis of an adequacy decision”. That is, whether the European Commission has decided that a country outside the EU/EEA offers an adequate level of data protection. To those countries, personal data may be transferred seamlessly, without any further safeguard being necessary, from the EU/EEA.

Inclusion of the US on that list is of course very important, not least for companies like Facebook and Google, and US companies offering cloud-based services as well. The Court of Justice of the European Union (CJEU) has deemed earlier transfer schemes (Safe Harbour and Privacy Shield) illegal, so “the whole world” is waiting for the EU Commission’s adequacy decision.

This came, as a draft, 14 February 2023 where the Commission concludes (page 54) that “… it should be decided that the United States ensures an adequate level of protection within the meaning of Article 45 of Regulation (EU) 2016/679, …)

(The figure below illustrates the “road” for legislative decisions in the EU. A more comprehensive description of the legislative procedure can be found here.)

However, the same day, 14 February 2023, the Committee on Civil Liberties, Justice and Home Affairs of the European Parliament concludes ” .. the EU-US Data Privacy Framework fails to create actual equivalence in the level of protection; ..” and “..urges the Commission not to adopt the adequacy finding;”.

Incompatible legislative frameworks

There are two important arguments, among others, behind the Commission’s conclusion: 1) There is no federal privacy and data protection legislation in the United States, and 2) the EU and the US have differing definitions of key data protection concepts such as principles of necessity and proportionality (for surveillance activities etc.), as pointed out by the Court of Justice of the European Union (CJEU).

Shortly thereafter, on 28 February 2023, the European Data Protection Board (EDPB) made public their opinion on the decision of the EU Commission regarding the adequacy. The EDPB has some concerns that should be clarified by the Commission, for instance relating to exemptions to the right of access, and the absence of key definitions.

Furthermore, the EDPB remarks that the adequacy decision is only applicable to US organizations which have self-certified, and that the possibility for redress provided to the EU data subjects in case of violation of their rights is not clear. “The EDPB also expresses concerns about the lack of a requirement of prior authorization by an independent authority for the collection of data in bulk under Executive Order 12333, as well as the lack of systematic independent review ex post by a court or an equivalently independent body.”, as stated in Opinion 5/2023.

The next step in the process is voting over the Commissions proposal in the European Parliament, probably in April, and thereafter the adequacy decision must be approved by all member states, before the EU Commission’s final decision.

The Commission may set aside the results of the voting in The Parliament, but one should expect that the critics from The Committee on Civil Liberties, Justice and Home Affairs, and the concerns of EDPB, will impact the implementation of the Framework.

Here it would be prudent to recall the statement made by the Austrian non-profit organization NOYB, chaired by Maxmillian Schrems: “At first sight it seems that the core issues were not solved and it will be back to the CJEU sooner or later.”. This refers to the verdicts of the CJEU (Court of Justice of the European Union) condemning the former frameworks Safe Harbour and Privacy Shield – the verdicts bearing the name Schrems I and Schrems II, respectively.

Bottom Line: The final outcome of the process is unclear, but in any event we have to wait for the final decision of the EU Commission.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought regarding any specific circumstances.

Continue Reading →

The Norwegian Consumer Council’s voice is heard worldwide

The Norwegian Consumer Council (NCC) has taken a strong position against commercial surveillance online, and has made it very visible how the Ad-Tech industry is exploiting personal data for business purposes.

“Big data” has since the entry of social platforms on the Internet, been accumulated and used unscrupulously by some companies for profit. Some of the players in the field are sharing information they collect on users with third party advertisers without their users’ knowledge or consent. The driver is all the money connected to targeted advertising. However, sharing of personal data in this way is prohibited according to the EU’s General Data Protection Regulation (GDPR).

The NCC has no authority to enforce personal data legislation, but the Norwegian Data Protection Authority (NDPA) does. And so, the NCC can freely report findings of breaches of the GDPR and Norwegian data protection regulations to the NDPA.

NCC and NDPA at the forefront

A good illustration of this interaction is the case against Grindr. Earlier this year the NCC, based on the report “Out of Control” (2020), raised the case against Grindr and five Ad-Tech companies that were receiving personal data through the app: Twitter`s MoPub, AT&T’s AppNexus, OpenX, AdColony and Smaato.

All the complaints were filed (in cooperation with the European Center for Digital Rights, noyb.eu), at the NDPA because of violations of the GDPR. The complaints concern Grindr transmitting sensitive personal data as for example group affiliation, sexual orientation, and geographic location, with several other parties without encrypting the traffic.

Even if data is anonymised, such as when third parties operate with their own proprietary identification numbers, it is possible to combine data from various sources with openly available information to produce a picture that can identify an individual.

In January, the NDPA announced a fine of 65 mill NOK (€ 5,8 M or approximately $ 6,5 M) on Grindr. The NCC has also in May this year acted against 8 companies and asked for details of their surveillance through the services Perfect365 and MyDays.

The Norwegian urge to protect personal data was also illustrated in May 2021. Then the NDPA submitted an advance notification of an administrative fine of NOK 25 mio to Disqus Inc. The company does widget tracking, analysing and profiling, and disclosing personal data to third party advertisers, and in doing so violates multiple articles (i.e. Article 6 and Article 7) of the GDPR.

Update on 30 September 2023 on the Grindr case

Grindr appealed the decision to the Norwegian Personal Protection Board, which rejected the appeal in its decision on 29 September 2023, and announced that it upholds the fine. – We have received the decision and cannot comment on it until we have been able to discuss it with our client, says the lawyer who represents Grindr.

The privacy movement grows stronger

All of these cases illustrate the NCC mission, but the NCC is working from a broader perspective: To establish a broad, international movement towards surveillance-based advertising.

This movement got a push with NCC’s seminal report Out of Control (2020), which has received media coverage in more than 70 countries, included the US and Japan (see our previous blog post).

Recently (June 2021), the NCC released another report: Time to ban surveillance-based advertising, with the subtitle The case against commercial surveillance online.

On page 4 there is quite a good summary of what the driving force is:

…today’s dominant model of online advertising is a threat to consumers, democratic societies, the media, and even to advertisers themselves. These issues are significant and serious enough that we believe that it is time to ban these detrimental practices.

In a coalition with more than 60 organizations from Europe and the US, including some 10 consumer organisations and the umbrella organisation BEUC – the European Consumer Organisation – the NCC on June 23 2021 sent an open letter to EU and US policymakers. The letter urges the policymakers to “…take a stand and consider a ban of surveillance-based advertising as part of the Digital Services Act in the EU, and for the U.S. to enact a long overdue federal privacy law.” The coalition is backing up its call with the reports by NCC.

On behalf of NCC, the consumer research company YouGov conducted a survey among a representative selection (internet population) 18 years+ about their attitude to surveillance-based advertising. The result was unambiguous: Only 10% responded positively to the idea of commercial actors collecting information about them online, and only one in five think that ads based on personal information is OK

Runbox has a clear standing against the collection of consumer data and surveillance-based advertising: Our service is ad-free, and we never expose our customers’ data for commercial purposes. We are very strict when law enforcement authorities in Norway or foreign countries request that we disclose data about our customers.

At Runbox we are proud to reside in a country that puts privacy first, and we wholeheartedly support the appeal to ban surveillance-based advertising. Therefore Runbox will annually donate to support noyb.eu, and we have joined the list of individuals supporting the appeal.

Continue Reading →

Annual Message 2021 from the Board of Directors

Dear customers, business partners, and shareholders,

Upon the completion of the Annual General Meeting in Runbox Solutions for the fiscal year 2020 we take the opportunity to review our company’s status in accordance with our commitment to transparency.

Although 2020 was in general a difficult and challenging year due to the corona pandemic and the worsening climate change, we celebrated the 20th anniversary of the Runbox email service in October with special subscription offers to our loyal customers. This, together with continued growth in our customer base and favorable currency exchange rates, resulted in a record year for Runbox financially.

Through the year we made significant progress on Runbox 7 development, and reinforced our commitment to privacy, security, and the environment.

Runbox 7 Developments

During 2020 we continued to improve and expand the groundbreaking Runbox 7 webmail application, which features search capabilities in the browser that provides immediate searching and listing of email. Runbox 7 combines the unique database-accelerated Runbox architecture with cutting-edge technologies such as WebAssembly, HTML 5 Canvas, and Progressive Web Apps to create an immediate email experience.

Developments have focused on expanding Runbox 7 toward a complete web application, and numerous enhancements have been made to Mail, Contacts, and Calendar. Additionally we have added innovative new features such as the Mail Overview and Popular Recipients, which provide a new level of message overviews based on sender and recipient data.

Runbox 7 development can be followed on our Runbox 7 Roadmap in the Runbox Forum and aims to solve the growing challenges of email interfaces and bringing forth the future of email.

The project is partially funded by the Research Council of Norway as a research and development (R&D) project in support of the innovative aspects of the solution.

Environmental Engagement

The operation of all business continues to depend on the foundation of our natural environment, which is under increasing pressure from human activity. In 2020 the world has seen a continued growth in greenhouse gas emissions with subsequent increases in extreme weather, wildfires, and droughts around the globe.

As inhabitants of Earth we all share the responsibility to decrease our negative impact on our environment. Runbox is built with a strong ethical foundation and we are dedicated to decreasing our ecological footprint and other environmental impacts that result from our operations.

The environment is a primary consideration when developing our services, and in 2020 we continued strengthening our commitment to having a positive ecological impact.

The data center where our email servers are located is 100% hydro-powered, and the electricity powering our email architecture is utilized exclusively by the Runbox email service.

In 2020 we achieved CO2 double negative operations by implementing our Environmental Policy and supporting the organization World Land Trust. In 2021 we have extended our commitment by partnering with the Norwegian tree-planting organization Trefadder, which creates and nurtures climate forests in Norway.

Renewed Commitment to Privacy and Security

Through the year we renewed our commitment to privacy and security, and the GDPR in particular. Our GDPR implementation has continued with reinforced policies, procedures, and technologies, and as a company located in Norway our service can rely on the strong Norwegian privacy protections.

All user data processed through the Runbox email service is stored on our own physical servers in Norway, and last year we completed the transition to encrypted SSD storage for all email account data.

We have continued our blog post series outlining our road towards GDPR compliance, and made further progress with privacy and security improvements to our services.

Growing our operations

We are working closely with our system management partner Copyleft Solutions to scale our email service infrastructure with a distributed system architecture to support the continued growth of our customer base.

Together with our development partners Shadowcat Systems and Peregrine Computer Consultants Corporation, our diverse team includes members from Norway, the UK, Poland, Brazil, and the US. The background, geographical location, and diversity of our team combined with a steadfast commitment to the ethics and policies of our company forms the core of our organization.

The contributions from our open source community on Github increase the security and speed of Runbox 7 development further, and we are excited to continue the race to revolutionize email in 2021.

Continue Reading →

Out of Control: Apps that share personal data revealed by the Norwegian Consumer Council

“If you are not paying for the product, then you are the product”.

This is a common saying when referring to online services that are offered for no financial payment (“free”).

The reason is that they often collect some personal data about you or your use of the service that the provider then can sell to the online advertising and marketing industry for payment. The payment they get for this covers the cost of providing the service to you and also allows for a profit to be made.

And so, they earn their money, and the app users are their product.

Apps as a source for big personal data

At Runbox we collect only the data that is required in order for us to provide our services to you, and that data is never shared with anyone for marketing or financial purposes.

However, it is common knowledge that companies like Google and Facebook use our personal data for targeted advertising. The personal data collected is anonymized and often aggregated to produce larger data sets, which enable them to target individuals or groups based on common preferences — for instance that they live in a certain location or like to drink coffee.

The idea that your data is anonymized might provide some comfort. But because of smartphones and the smartphone software applications (“apps”) many people use, companies can collect a large range of types of data and so trace individuals without asking for personal details such as your name. An example of this type of data is your smartphone unique identifier (IMEI-number1), and IP-address (when connected via WiFi).

Combined with your email address, GPS data, app usage etc., it is possible to identify specific individuals -– namely you!

Exposing the AdTech industry

To investigate this issue, The Norwegian Consumer Council (NCC), a government funded organization representing consumer interests in Norway, published a groundbreaking report last year about how the online marketing and AdTech (Advertising Technology) industry operates.

The report’s title immediately raised the flag: “Out of Control” (OuC)2. And the subtitle outlines the findings: “How consumers are exploited by the online advertising industry”.

The report tested and analyzed 10 popular “apps” under the umbrella “social networking apps”, and the findings were concerning. Most users of such apps know that registering your personal data is optional, and after the introduction of the GDPR every app is careful to ask for your consent and encourages you to click OK to accept their Privacy Policy.

What many users will not know is how much and how far the personal data is distributed. Only a few users will be aware that clicking OK implies that your data is fed into the huge AdTech and MarTech industry, which is predicted to grow to USD 8.3 billion in annual revenues by 20213.

The players in this industry are giants such as Amazon, Facebook, Google and Twitter. If that was not enough, both iOS (Apple) and Android (Google) have their ways to track consumers across different services.

Apple being more privacy minded than some others have recently developed options to allow the user to reset the “unique” advertising identifier in devices and also stop tracking across WiFi networks to break the identification chain and make it harder to target a specific user.

But the industry also has a large number of third-party data and marketing companies, operating quietly behind the scenes.

The far-reaching consequences of AdTech

This is what the NCC’s report is about, and the findings are concerning:

The ten apps that were tested transmit “user data to at least 135 different third parties involved in advertising and/or behavioral profiling” (OuC, page 5).

A summary of the findings is presented on OuC page 7, and here we find social networking apps, dating apps and apps that are adapted to other very personal issues (for instance makeup and period tracking). The data that is gathered can include IP address, GPS data, WiFi access points, gender, age, sexual orientation, religious beliefs, political view, and data about various activities the users are involved in.

This means that companies are building very detailed profiles of users, even if they don’t know their names, and these data are sent to for instance Google’s advertising service DoubleClick and Facebook. Data may also be sold in bidding processes to advertising companies for targeting advertising.

It is one thing to see ads when you perform a Google search, but it’s quite another to be alerted on your phone with an ad while you are looking at a shop’s window display, or passing a shop selling goods the advertiser knows you are interested in. Scenarios like these are quite possible, if you have clicked “OK” to a privacy policy in an app.

Personalized directed ads are annoying, but even worse is that the collection and trade of personal data could result in data falling into the hands of those who may then target users with insults, discrimination, widespread fraud, or even blackmail. And there is clear evidence that personal data have recently been used to affect democratic elections4.

What happened after The Norwegian Consumer Council published “Out of Control”, will be covered in our next blog post, but we can reveal that one of the companies studied had a legal complaint filed against them for violating the GDPR and is issued an administrative fine of € 9.6 million.

So stay tuned!

References:

  1. IMEI stands for International Mobile Equipment Identity.
  2. The report Out of Control was referred to in our previous blog post GDPR in the Wake of COVID-19: Privacy Under Pressure.
  3. Source: https://privacyinternational.org/learn/data-and-elections
  4. Source: https://bidbalance.com/top-10-trends-in-adtech-martech/

Continue Reading →

The Norwegian COVID-19 contact tracing app is banned by the Data Protection Authority

GDPR in the Wake of COVID Spread: Privacy under Pressure – Part 2

Our previous blog post in this series concerned mobile phone applications under development, or already developed, in various countries for tracing the spread of COVID-19 infections. In particular the blog described the situation in Norway, and we expressed our concerns, but also our trust, in the fact that The Norwegian Data Protection Authority (‘Datatilsynet’) would be on the spot to safeguard privacy – as regulated by strict Norwegian privacy regulations.

The Norwegian Data Protection Authority — more than a watchdog

Temporary suspension of the Norwegian Covid-19 contact tracing app
The Norwegian Smittestopp app

We were right, and we are proud of the intervention by the Norwegian Data Protection Authority (NDPA), which in June banned the Norwegian COVID-19 tracker app Smittestopp. The ban illustrates NDPA’s independency, and that NDPA has legal power to enforce privacy protection when public (and private) organizations violate the law.

This power is anchored in the Personal Data Act (personopplysningsloven), the Norwegian implementation of GDPR, and the Personal Data Regulations (personopplysningsforskriften).

After evaluating the app Smittestopp as it was implemented in April this year, NDPA concluded that the app violated the privacy legislation in mainly two respects:

  1. The app was not a proportionate intervention of the user’s fundamental right to data protection.
  2. The app was in conflict with the principle of data minimization.

On June 12, The NDPA notified The Norwegian Institute of Public Health (NIPH) that the app would be banned, which was confirmed on July 6. Consequently, NIPH immediately stopped collecting data from the around 600,000 active users of the app, and deleted all stored data on their Azure server.

What the requirement for proportional intervention means

The breach of the requirement for proportional intervention concerned the expected low value of the app regarding infection tracking, due to the relatively small number of the population in the testing areas actually using the app (only 16%).

The reason for the breach of the principle of data minimization was that the app was designed to cover three different purposes:

  1. Movement tracing of individuals (for research purposes).
  2. Spread of the infection among the population.
  3. The effectiveness of infection control measures.

The NDPA was also critical to the app because it was not possible for the users to choose for which of the three purposes their data would be used.

A new app is already being planned

The government has decided to terminate further development of Smittestopp, and will instead focus on the development of a new app. After seeking advice from NIPH, the government has decided to base a new app on the Google Apple Exposure Notification (GAEN) System, or ENS, which they call “the international framework from Google and Apple” because many countries (for instance Denmark, Finland, Germany, Great Britain) are going “the GAEN way”.

Important arguments for the government’s decision are that GAEN supports digital infection tracking only (Bluetooth-based), involves no central data storage, and includes the possibility to exchange experiences and handle users’ border crossings. In the meantime the EU has implemented a recommendation for decentralized Corona tracking applications, putting GAEN “squarely in the frame“.

NIPH was given the task to specify a request for proposal in an open competition for the development assignment of the new app, and now (October 20) the Danish Netcompany is hired to do the development. Netcompany has a similar contract with the Danish health authorities, and was the only bidder (!). The new app expected to be implemented this year (2020).

The privacy debate continues

Three main issues are still being debated, and the first is technical: Is Bluetooth reliable enough? Experiences show that false positives, but also false negatives, do occur when Bluetooth is being used.

The second issue is of course privacy. Even if personal data is stored locally on the phone, notifications between phones have to be relayed through a network – so what about hacking? In addition, Trinity College in Dublin has uncovered that on Android phones, GAEN will not work unless it is sending owner and location information back to Google.

This leads to the third issue: Is it sensible to let the tech giants control a solution that involves processing very personal information? “Do Google or Apple get to tell a democratically elected government or its public health institutions what they may or may not have on an app?”

The Norwegian Data Protection Authority published a report on digital solutions for COVID-19 (‘Coronavirus’) infection tracking on September 11 this year. The report was developed by Simula Research Laboratory, who did not bid on the contract for the new GAEN-based application (arguing that they are a research institution and not a software development company).

The report “… focuses on efficiency, data privacy, technology-related risks, and effectiveness for government use. In terms of privacy and data protection, the report notes that if location data is still stored by Google, the COVID-19 app Smittestopp would be less privacy intrusive than the GAEN one.”

Conclusion

We will conclude with a quote (in our translation): “There is no perfect solution for digital infection tracking. Effective infection control and privacy stand in opposition to each other.”

For us at Runbox, privacy is priceless, and we are still wondering if the pros outweigh the cons.

Continue Reading →

Data Privacy Day

January 28th is Data Privacy Day, and was initiated by the Council of Europe in 2007. Since then, many advances to protect individuals’ right to privacy have been made.

The most important of these is the European Union’s General Data Protection Regulation (GDPR) which was implemented on May 25, 2018. Runbox has promoted data privacy for many years, anchored in Norway’s strong privacy legislation.

At Runbox, which is located in the privacy bastion Norway, we believe that privacy is an intrinsic right and that data privacy should be promoted every day of the year.

Your data is safe in the privacy bastion of Norway

We’re pleased that Data Privacy Day highlights this important cause. Many who use the Internet and email services in particular may think they have nothing to hide, not realizing that their data may be analyzed and exploited by corporations and nation states in ways they aren’t aware of and can’t control.

While threats to online privacy around the world are real and must be addressed, we should not be overly alarmed or exaggerate the problem. Therefore we take the opportunity to calmly provide an overview of Norway’s and Runbox’ implementation of data privacy protection.

Norway enforces strong privacy legislation

First of all, Norway has enacted strong legislation regulating the collection, storage, and processing of personal data, mainly in The Personal Data Act.

The first version of Norway’s Personal Data Act was implemented as early as 1978. This was a result of the pioneering work provided by the Department of Private Law at the University of Oslo, where one of the first academic teams within IT and privacy worldwide was established in 1970.

Additionally, the Norwegian Data Protection Authority, an independent authority, facilitates protection of individuals from violation of their right to privacy through processing of their personal data.

For an overview of privacy related regulations in the US, in Europe, and in Norway, and describes how Runbox applies the strong Norwegian privacy regulations in our operations, see this article: Email Privacy Regulations

Runbox enforces a strong Privacy Policy

The Runbox Privacy Policy is the main policy document regulating the privacy protection of account information, account content, and other user data registered via our services.

If you haven’t reviewed our Privacy Policy yet we strongly encourage you to do so as it describes how data are collected and processed while using Runbox, explains what your rights are as a user, and helps you understand what your options are with regards to your privacy.

Runbox is transparent

Runbox believes in transparency and we provide an overview of requests for disclosure of individual customer data that we have received directly from authorities and others.

Our Transparency Report is available online to ensure that Runbox is fully transparent about any disclosure of user data.

Runbox is GDPR compliant

Runbox spent 4 years planning and implementing EU’s General Data Protection Regulation, starting the process as early as 2014.

We divided the activities implementing the GDPR in Runbox into 3 main areas:

  • Internal policies and procedures
  • Partners and contractors
  • Protection of users’ rights

This blog post describes how we did it: GDPR and Updates to our Terms and Policies

Runbox' GDPR Implementation

More information

For more information about Runbox’ commitment to data privacy, we recommend reviewing the Runbox Privacy Commitment.

Continue Reading →