Communication

mattiealston Tue, 12/15/2020 - 00:00

Topic

Problem:

In order to understand the likelihood of success, the user wants to know how effective the algorithm is before initiating an operation.

The "PictureThis" app boasts about its 95% accuracy rate, which is also implicitly being honest about its 5% failure rate.

Solution:

The system communicates an effectiveness metric. For example, in pattern recognition operations this could be indicated in terms of what percentage of image capture inputs result in a successful match with the correct data points.

Discussion:

Obviously if a system communicates that it is less than 100% successful 100% of the time, that can undermine the confidence a user has in that system. But in the the long term, acknowledging the fallibility of a system is a vital part of building trust in it (as per Setting Expectations and Acknowledging Limits).

Quantifying the effectiveness with a score provides a couple of advantages over a vaguer, copy-driven approach. Firstly, if the metric is comparable to ones used by similar competitor systems, the user can easily establish how this system compares to others. Secondly, as the score is subject to change, this allows the system to track and communicate how it has improved over time — acknowledging initial shortcomings while establishing a trajectory towards increased success encourages the user to return later and engage further, as opposed to simply overpromising and underdelivering on first use, which can discourage further exploration.

Apologies for Inaccuracy

leighbryant — Thu, 15 Oct 2020 00:00:00 +0000

Apologies for Inaccuracy

Application

leighbryant Thu, 10/15/2020 - 00:00

Topic

Problem:

The AI in a system may generate inaccurate results that cause confusion, may be offensive, or otherwise unsettle the user. The user would like to know that this is a possibility and feel comfortable with the system regardless.

Microsoft's "How Old" AI-powered software guesses the age of the individual featured, but also includes an apology disclaimer for inaccurate guesses beneath the image.

Solution:

The system apologizes for any possible inaccuracy that may be present, especially in regards to sensitive topics that could cause harm or offence.

Discussion:

While it is important for systems to admit to their fallibility (via Setting Expectations & Acknowledging Limitations, for example), there is an additional requirement to address how this fallibility may affect the user. In reality, what we call "the user" is a million unique individuals each of whom want to be treated with dignity and respect, not just in this interaction but in every facet of everyday life. We need to use different lenses, with more expansive understandings of who "the user" is in terms of social, psychological, physical, and ideological variables that describe any one embodied individual that goes beyond standard task completion assessments. Apologizing for potentially harmful interpretations acknowledges the humanity of the individuals.

Automated Subtitles

mattiealston — Wed, 15 Jul 2020 00:00:00 +0000

Automated Subtitles

Application

mattiealston Wed, 07/15/2020 - 00:00

Topic

Problem:

Users want to know what is being said in video content. The system designers want to facilitate this without manually transcribing all videos.

While a manually created transcription might be more accurate, YouTube's automated subtitling is very effective at making videos accessible for users.

Solution:

Speech recognition automatically generates subtitles for video content. The user can choose to activate the subtitles as required.

Discussion:

Similar to Automated Alt Text, the benefits of this pattern are more in the efficiency gains to the designers than an improvement in the experience of the end users, who care more about the existence of subtitles than their source.

That said, we shouldn’t overlook the empowering potential of patterns like this — it isn’t that the users should be wowed by an AI automagically producing content, but rather that often the deployment of automation functionality like this can mean the difference between the content existing or not. In the case of automatically generated subtitles for video, this can make previously inaccessible content available to users with hearing impairments, which is of immense value to them.

Automated Text Summarization

mattiealston — Mon, 15 Jun 2020 00:00:00 +0000

Automated Text Summarization

Application

mattiealston Mon, 06/15/2020 - 00:00

Topic

Problem:

In situations where there is a lot of body text, users with cognitive or literacy issues want to know what is contained within that content without having to read the whole text.

The Automatic Text Summarizer tool lives up to its name, demonstrating how well this functionality can work.

Solution:

The user can click to view a shortened and simplified version of the text that has been generated by an AI.

Discussion:

While patterns such as Automated Subtitles are certainly empowering, they are really just instances of automating the production of conventional content, and do not go far in demonstrating the true revolutionary potential of AI in this domain. Automated summarization, on the other hand, shows that AIs can actually produce valuable original content, and to varying specifications to match users’ unique needs.

Yes, it would be possible for a copywriter to simply create two or three different versions of a text to suit different reading levels, but this is labour intensive and rarely done. With an AI automating this process, the opportunity is there to allow for multiple versions of the same text with little hassle, opening the door for true personalization that fits each user’s own abilities and needs.

Explanation of Algorithm

leighbryant — Fri, 15 Feb 2019 00:00:00 +0000

Explanation of Algorithm

Application

leighbryant Fri, 02/15/2019 - 00:00

Topic

Problem:

The user wants to know, in general terms, how the AI is operating: what it takes as inputs, what calculations it runs, and what it outputs.

A GAN-created image with a pop-out text box providing the source of the image as well as links explaining how this image was created.

Solution:

Either within the app or as a separate content area accessed via a link, there should be a concise but complete description of the AI's workings. It should be easily understood by lay readers and not obscure details through vague language, highly technical jargon, or misleading metaphors.

Discussion:

The craving to understand how things work, while not manifested in every user, is fundamental to the human condition. In some cases, users want to understand so they can determine how to drive the result they’d like to see from the algorithm. Whatever the motivation for the user, providing an explanation is more than an exercise in building trust (though it is that, too)— it can also be an opportunity to delight by satisfying a recreational curiosity or offering the user additional control.

More Examples:

Microsoft's CaptionBot.ai provides a wealth of information about how it works behind the scenes.

More image recognition-based AI applications provide users with additional information on how they work.

Gender Neutral Voice

mattiealston — Thu, 15 Nov 2018 00:00:00 +0000

Gender Neutral Voice

Application

mattiealston Thu, 11/15/2018 - 00:00

Topic

Problem:

The user wants to interact with a voice interface which uses a voice that isn’t distinctly male or female.

The pioneering Q voice is designed to sound neither female nor male.

Solution:

The voice used by the interface is carefully constructed to sound neither male or female.

Discussion:

As with a Gender Neutral Bot, the perceived gender of a voice UI does not affect task completion and so may be overlooked. However, unlike bots which use interface, text, graphics, and emojis to communicate, the voice literally is the interface for a voice UI, so our entire perception of that interface is determined by the tone, expression, and vocabulary of it. The trend has been to default to female voices in these interfaces, which is especially problematic given that they often play the role of domestic servants or are otherwise subservient.

There are least two benefits to using a gender neutral or genderless voice like Q. One is immediately clear, that of avoiding the perpetuation of harmful stereotypes. The second one may be less obvious, but is also worth considering — that a significant number of people identify as non-binary or otherwise reject binary gender labels. Providing a non-binary voice option is a valuable recognition that people with non-binary genders exist.

Risk Alert

leighbryant — Thu, 15 Jun 2017 00:00:00 +0000

Risk Alert

Application

leighbryant Thu, 06/15/2017 - 00:00

Topic

Problem:

Users want to know how much trust to place in the system's intelligence, especially in critical applications such as the diagnosis of health issues, where there can be real world implications of following bad advice.

A medical app alerts users to the importance of seeking additional input from a professional.

Solution:

In combination with Setting Expectations & Acknowledging Limitations, the system alerts the users of the risks they may encounter in following its advice, and how to mitigate that risk (i.e. to seek further advice from an expert before proceeding with any action).

Discussion:

While many faulty AI-driven predictions will result in little more than an ill-fitting dress or unliked movie, some are far more impactful. Great care should be taken in addressing the risks to the user, both because there can be potential legal and commercial implications, and because there is a moral imperative to reduce potential harm where possible.

Setting Expectations & Acknowledging Limitations

leighbryant — Sat, 15 Apr 2017 00:00:00 +0000

Setting Expectations & Acknowledging Limitations

Application

leighbryant Sat, 04/15/2017 - 00:00

Topic

Problem:

Users want to know what the AI being used in the application is actually capable of.

By clarifying the parameters of the outputs the AI application is providing, the system sets clear expectations for the users in the quality of the results.

Solution:

Before initiating a calculation or displaying results, the system tells the user what the AI can calculate, what the limits are in terms of accuracy in the output, which outputs are unlikely, and what is impossible to achieve that the user might otherwise want.

Discussion:

Trust is built on transparency, which necessarily includes communicating the limitations of the AI. Therefore this is partly about managing the user’s expectations throughout the process, but also about setting an appropriate tone of voice and brand position, one based on honesty and humility.

Anti-Pattern: Hijacked Chatbot

mattiealston — Mon, 15 Jun 2015 00:00:00 +0000

Anti-Pattern: Hijacked Chatbot

Application

mattiealston Mon, 06/15/2015 - 00:00

Topic

Problem:

While an AI chatbot learns from users en masse, some of those users want to bias that AI towards using offensive language and repeating bigoted opinions. While these users are entertained by their perverse creativity, others find the resulting behaviour repulsive.

Microsoft Tay on first launch. Within hours, Tay turned from a playful innocent into a Nazi troll.

Anti-pattern response:

A chatbot released into the wild learns from all users and adds the collected knowledge and patterns of speech into its vocabulary. A determined subset of users apply an organized collective effort to feed the chatbot offensive data that it will later repeat in various ways. As the chatbot becomes increasingly notorious, more trollish users are attracted towards the project and contribute to it, leading to a negative feedback loop and the quick decline of the bot into unsalvageable profanity.

Discussion:

The archetypical case study for this is of course Microsoft's Tay from 2016, a chatbot that utilized a Twitter account to communicate with the world. Tay initially had the mannerisms and personality of a teenage American girl, but within 16 hours the account had become a vile mess of sexist and sexually explicit content, racist memes and offensive language, and was quickly shut down amidst a PR disaster. Some of this was the result of simple hijacking— a “repeat after me” function allowed users to rebroadcast whatever they wished. Some, however, was genuinely learned from the users and integrated into the AI’s behaviour, making it impossible to eradicate or correct.

The lessons are simple, if unpleasant to acknowledge: if you release an AI that is capable of learning from everyone indiscriminately with no control or moderation over the input data, it can be exploited by bad faith actors. And given the propensity of the internet to act as a catalyst for trollishness and a network for bigots to organize effectively, if that exploitability is discovered and shared, it is inevitable that it will quickly be put to the worst possible use. The rapidity and ferocity of Tay’s meltdown might be remarkable, but any AI designer should expect similar effects from similarly unrestrained bots.

Anti-Pattern: Insensitive Bot Dialogue

leighbryant — Fri, 15 May 2015 00:00:00 +0000

Anti-Pattern: Insensitive Bot Dialogue

Application

leighbryant Fri, 05/15/2015 - 00:00

Topic