Paul Krizsan | think moto

Connecting with Microsoft AI and NVIDIA in Berlin

Posted on 20 März 202423 Juni 2025 by Paul Krizsan

When Microsoft announced that they would be in the neighborhood with one of our favorite topics, we knew we had to attend the Microsoft AI Tour stop in Berlin. With an agenda full of fascinating topics it was hard to choose which sessions to attend for our technology and AI team.

The tenor of the event has been one of reliability and maturity. Microsoft and their event partner NVIDIA wanted to let everyone know that artificial intelligence is long past the experimental stage of proof-of-concepts and haphazardly assembled tech demos. AI has reached a point wherein scalability and establishment of best practices and operations is now more critical than sheer feasibility.

In this post we have broken down four of the key areas and what we have observed and learned.

Strong commitments to security and trust

„How many of you trust AI?“ started the speaker, and the answer became evident upon counting the hands in the audience–not many. Most AI users usually don’t really know what is happening under the hood, and it is difficult to predict the output. Even the researchers and scientists behind the technology are constantly discovering new behaviors and peculiarities of their models, not seen before.

As we all know, LLMs are usually trained on data from across the internet, thus prompting the clear question: What do they know about me, and if they will use what I said when talking to other users?

At this point, we all have many questions about our privacy when using AI. Therefore, Microsoft has unveiled the curtain behind their approach to securing AI and has shared what should be taken into account when building an AI-powered application.

At their core, LLMs and foundation models that have not been fine-tuned are difficult to trust: unpredictable hallucinations and with few guard rails to reign in their output.

However, using additional layers of safety on top of the foundation models, such as grounding it with high-quality knowledge bases and self-evaluating patterns and prompt instructions, developers can define safe corridors for reliable and consistent output and responses.

Safety is a critical component that stretches across the entire bandwidth of generative AI: from the data ingested to instructions during processing of input and output, all the way to disclaimers and labels in the user interface.

So our question should be reframed: Do I trust the developers behind the specific AI tool?

Apart from discussing general concepts and best practices, Microsoft has introduced its new Security Copilot, which will be available from May 1st and can function as a SOC Analyst. It can answer questions about the state of your Microsoft apps connected to the cloud, conduct security audits, and assist in debugging various application errors. Moreover, it can even alert you if an email you receive contains an attempted phishing attack.

Copilots, not autopilots

The vibrantly enthusiastic Seth Juarez summed it up perfectly: the current suite of AI that Microsoft can provide is not meant to run all on its own. You need to assist it in refining the desired output, guide it and tell it what worked and what maybe is not correct.

With multiple breakout and workshop sessions on how to build your own copilot throughout the day being completely overrun and people being turned away from overflowing rooms, it was obvious what topic is on everyone’s mind.

From copilots that monitor your systems and logs to report on potential vulnerabilities and breaches to copilots that summarize analyze business data and create reports with suggestions. The possibilities are endless and nearly every Microsoft product is now equipped with it’s own copilot feature.

But what is clear is that this plethora of different entry points into conversational interfaces also seems convoluted and raw. UX patterns and output formats vary greatly and the question arises: wouldn’t it be a lot more accessible and friendly to have a singular interface with the AI rather than having to jump between them constantly?

A better solution will likely be OS-integrated LLMs that provide standardized patterns or modules that application can tap into, but we are not quite there yet. Until then, every service and product will likely have its own standalone conversational interface before they will be consolidated in a swift move on the OS level.

RAG and AI search

Vector search and retrieval-augmented generation (RAG) still has generative AI firmly in its grasp. The capabilities of vectorized databases for grounding are profusely necessary for reliable output and a significant reduction of hallucinations.

Both Microsoft and NVIDIA presented models and best practices for workflows for preprocessing and referencing of vector databases as well as demonstrations for Azure AI studio.

While it seems as though technological advancements have slowed a bit in this field recently, there were some takeaways for us that illustrated the necessity for indexing when handling large amounts of documents for the embedding model.

Establishing LLMOps for production-ready applications

What fell short in my eyes was the sharing of best practices on how to enable less tech-savvy people in organizations and provide building blocks for prompts or the development of pipelines of building blocks for prompts.

One of the biggest challenges of establishing acceptance of LLMs and AI within organizations is the fact that, while we are communicating on an intuitive level, natural language interfaces are not something we are accustomed to. Learning how to properly phrase prompts with their intransparent intricacies of what you should include and how, should be managed and accompanied with tooling that steers users accordingly.

A substantial part of building the right infrastructure should be the automated inclusion–or at least privision–of repeatedly and frequently used excerpts and prompt modules that get attached to prompts. If I am generating content for my marketing materials, I should not constantly have to paste instructions for the tone of voice of the output or camera settings for consistent image generation.

While Microsoft covered potential workflows for rewriting user queries before generating the final output, a lot of struggles and challenges of these enterprise-level ops management and tasks was unfortunately not mentioned.

In conclusion

Overall we greatly enjoyed the event, meeting fascinating people (some with the obligatory and adorned Apple Vision Pro) and listening to what drives and motivates others in the realm of artificial intelligence these days.

Thank you Microsoft and your efforts in outreach to developers and advocates!

What ChatGPT and LLMs Mean for How We Build Conversational Interfaces for the Future

Posted on 28 Februar 2023 by Paul Krizsan

Firstly, what are LLMs and ChatGPT? This is not an article about what LLMs (or large language models) and ChatGPT are. If you have been living under a rock and are unfamiliar with these names and terminology then this article written by ChatGPT explaining itself should be a good starting point.

We have been receiving questions from – and participated in many discussions with – our customers and peers about this exciting new tech and wanted to clarify our stance on where we see the opportunities and weaknesses at the current stage, as well as looking forward to a potential hybridized future. The biggest talking point has been the need for conversation design in an increasingly automated and generative world.

From our perspective as experts on conversational interfaces and conversation design we see predominantly two paths that this technology and trend will continue to develop on: the path of consumer-facing applications and the path of the technology as a tool and force multiplier. Neither of which will be eliminating the need for humans behind the wheel, steering the technology, anytime soon.

Hopping on the LLM bandwagon

Broadly speaking, this technology and its implications are spreading at breakneck speed. Many platforms are currently aiming at capitalizing on this goldrush-like state. You may have heard of Microsoft implementing ChatGPT in Bing and Google looking at fusing their proprietary equivalent LaMDa with their own search engine. These search engines follow a trend that companies such as SoundHound have been pursuing for a while, responding to users not in lists of search results, but in concrete answers in the form of natural language.

Other examples of quick wins in this brand new space are bot platforms such as Voiceflow and Cognigy.AI. Here the same purpose of applying LLMs to dynamically generate the system responses or predictable training data for intent training is being used heavily. Some platforms, like Cognigy.AI, are also considering going a step further and looking into the empowerment of conversation designers by allowing the creation of flows and elements through natural language prompts, speeding up the process of setting up new conversations greatly and thus contributing to rapid prototyping capabilities of these low-code platforms. Will these features collate into conversations that are production-ready, about to be rolled out to millions of users, out-of-the-box? Of course not. But they provide a good first framework to expand upon.

Trust in the system and the tech is dwindling

Widely broadcasted anecdotes of tech journalists and influencers, as well as hear-say from colleagues and friends have recently lead to a lot of skepticism when it comes to the current state of the technology. Articles quoting the unsettling feeling, individual erroneous responses and behavioral patterns reinforce negative connotations when it comes to LLMs in todays world. This obviously has a huge negative impact on consumer-facing applications.

Finding an appropriate place for LLMs should not be difficult

Focusing on this new technology as a force multiplies and enablement tool, is therefore the more stable path from our perspective. At least while the technology matures and new, more refreshing experiences for consumer-facing applications improve the publics perception in the mid-term.

On a more immediate and applied note, ChatGPT and LLMs are a great vehicle for innovation and a popular driver for change, but they are tools and will not replace human experts in conversation design. It is a good gap-filler and repetitive tasks but it will not provide the confidence and accuracy of dialogues designed by humans for a while.

The conversation designer is still the agent of change for this new tech

Our workflows in the future could consist of conversation designers laying down the structure of a dialogue, such as the starting point, the goal of the conversation and some checkpoints along the way, with the generative AI or LLM filling the gaps.

In an ideal world we would provide the AI with a purpose and a personality, but no actual dialogue would need to be written by humans. The conversation designer would be focused entirely on the strategic purpose of the interface and the decision on a vector of the personality and tone of voice of the bot.
Paul Krizsan, Director Conversational AI

So while remaining up to date with the current developments of this exciting new technology is vital, we do not share the current ubiquitous sentiment that users are ready for unfettered access to potentially image-harming experiences without having some of the kinks of current LLMs ironed out over the course of 2023.

Are you interested in talking about conversational interfaces, LLMs and how to design for conversations? Talk to us!

Visual power seeks identity–Impressions of CXI_19

Posted on 21 Juni 2019 by Paul Krizsan

Brick meets Brand

Now in its 11th year, Europe’s largest conference on corporate and brand identity is taking place. The venue for this prestigious event was once again the Bielefeld Lokschuppen, a building from the early 20th century with brick walls and a deeply industrial look.

The CXI is organized by the Fachhochschule Bielefeld, in particular the Faculty of Design. With this educational institution as the organizer, it is therefore hardly surprising that students are primarily represented in the registration and indeed in the audience.

Once again this year, six pairs of speakers and three-person constellations took to the stage to represent agencies and their clients in various brand projects. The aim was to create a unique, comprehensive picture of both parties involved in said projects. After all, otherwise you only get to hear from one side.

FOMO tv | Kurppa Hosk

The CXI_19 kicked off with the conference’s only international speakers. Swedish design agency Kurppa Hosk, represented by Thomas Kurppa, and Stockholm-based gallerist, artist, and jack-of-all-trades Jonas Kleerup shared the highly creative and nostalgic process of finding FOMO tv’s identity.

Kleerup’s idea of a video streaming platform on the topic of FOMO (Fear Of Missing Out) might be difficult to understand in its approach and of questionable relevance apart from the artistic approach, but the intensity and quality of the design left nothing to be desired. With impressive showreels of the brand-in-motion, a style reminiscent of VHS tapes and a flexible logo concept reminiscent of MTV’s myriad of logo variations, Kurppa Hosk overwhelmed the audience with visual power.

Kleerup’s idea of a video streaming platform on the theme of FOMO (Fear Of Missing Out) might be difficult to understand in its approach and of questionable relevance apart from the artistic approach, but the intensity and quality of the design left nothing to be desired. With impressive showreels of brand-in-motion, a style reminiscent of VHS tapes and a flexible logo concept reminiscent of MTV’s myriad of logo variations, Kurppa Hosk overwhelmed the audience with visual power.

To do so, Kurppa Hosk applied her own snowball-like design process to branding, approaching a finished brand gradually but with increasing speed and complexity. An interesting process, but more akin to the waterfall model than agile design, which made one wonder if this approach might be less suitable for more commercial projects due to inflexibility, at least on the outside.

Volkswagen | think moto

We had the honor to be on stage with one of our clients again this year as an agency. Between late 2017 and mid-2018, we broke new ground and into new realities with Volkswagen.

To bring Generation X closer to the experience of car dealerships and shiny new cars, Volkswagen developed a series of applications for virtual reality and mixed reality. The basis of the applications was the possibility to bring different car models to the users in a playful and virtual way. The only thing missing was the smell of new cars.

The involvement of think moto started with the dissonance of many interaction patterns within these prototype-like applications. After all, no one had yet addressed the coherence of interactions across different realities. As a rule, there were hardly any applications that had to pass the test of brand conformity in both virtual reality and augmented reality. Our mission, therefore, was to create a unified pattern library to keep existing and future extended reality applications in line with other Volkswagen digital products.

OSRAM Continental | KMS TEAM

When two people move in together, the household goods do not double. This was also the experience of Nadine Schian, Head of Communications, Marketing & Brand at OSRAM Continental. In a tough, three-year struggle, the joint venture between OSRAM and Continental built itself up on the basis of a technological partnership in the field of mobility lighting solutions and faced an important challenge in the process. As the brainchild of two down-to-earth and prestigious corporations, the task of finding its identity was either to follow in the footsteps of its parents or to break completely new ground.

This question was asked even before the actual founding of the joint venture, which for Vera Schnitzlein and the southern German agency KMS Team also meant: How do you build a brand without an existing company behind it? After all, nothing had been fixed yet, let alone signed.

Visually, therefore, basic elements of both brands were used. It was particularly important to find a color scheme for the colors that could stand alone but was derived from the parents.

For OSRAM Continental, however, the advantage of such an early start to finding an identity was above all that the corporation was able to launch on the first day after its official founding with around 1,500 employees on all continents and a fully developed, independent and expressive identity. A flying start for the young joint venture.

FC Bayern München | Interbrand

How do you tell an absolute Bayern Munich fan that his club’s favorite color is now yellow and green? You don’t. Philipp Mokrohs, the club’s lead brand strategist, and Alexandra Gövert of Interbrand told us about the problems of designing an incredibly emotional brand like the world’s most famous soccer club. In 2017, Gövert and her team were tasked with touching FC Bayern Munich’s identity and bringing it into the 21st century.

For a brand whose core values are all about tradition and heritage, change means putting on kid gloves for design. Major changes would have vehemently fallen victim to the emotions of millions and millions of fans, but at the same time Mokrohs reported a need for change. Evolution instead of revolution, was the motto.

The logo, for example, which hadn’t been touched in decades, was in desperate need of craft help and was first freshened up by Alexandra Gövert’s team. This was followed by the introduction of various key visuals, the gradient in the background, and the cut and contrast of fonts and images. In the end, it became clear that the company also needed its own typeface to ensure a consistent brand presence in the future. This way, Bayern Munich is not only secure on the pitch, but also on the web (and other touchpoints).

sipgate | g31

Agile, lean and in sprints to success. That’s actually a good recipe, isn’t it? That’s what the Internet telephony provider sipgate and the small Düsseldorf agency g31 thought when they tackled the rebranding of sipgate. Before they got to that point, however, the two partners had already taken a turn.

When sipgate originally came to g31, the talk was of a brand refresh – not a complete rebranding with a new identity. A project that was to be tackled in the classic waterfall approach. After the guys and gals from g31 around Mats Kubiak and Paul Schoemaker had then locked themselves away for two months and worked out a concept to meet this requirement, it was time to present.

However, the concept presented did not seem quite right for Tim Mois and Tobias Ritterbach from sipgate. Instead, they wanted something completely new. To achieve this, they now relied on two-week sprints, lively exchanges and employee surveys. Thus, week by week, the company’s stale body with its dotcom look grew into a sleek bolide in black and white with colorful facets for the Internet telephony provider’s many different products.

The previously almost independent and not at all visually consistent products and sub-brands were now captured and merged into a whole via patterns, uniform logos and a holistic identity system.

DHL | Strichpunkt

After DHL had already completed a tough and elaborate redesign of the brand a few years ago, there was no question of allowing the logo, the colors or even the key visuals to be touched, brought on board Strichpunkt to give the yellow logistics brand new vigor.

Thick tomes of style guides, separate online portals for digital and print media, and a plethora of different layouts and touchpoints from airplane wrappings to advertising posters in the DHL branch, each with its own set of rules, had slowed down the logistics giant and now threatened to be unnecessary ballast for an agile future for DHL.

Strichpunkt recognized this quite correctly and worked to consolidate the visual components of the brand. A UI toolkit for developers now helps keep new digital products consistent across different devices and shortens development time enormously. The two heavily text-heavy and opaque Brand Portals became one with lots of images, examples and help. The font selection, which previously consisted of over a dozen typefaces, was replaced by the new house font „Delivery“.

*DHL UI-Toolkit — Courtesy of Strichpunkt*

But probably the most impressive result from Strichpunkt was the creation of a layout generator, via which every DHL employee can easily create brand-compliant digital and print publications and products thanks to templates and guidelines. Just fill in the text fields, select images from the online library and you have a print-ready PDF in the highest quality and conforming to the brand from the choice of colors to the spacing.

Strichpunkt made all these changes under the concept of simplification. There should be flexible and universal principles and elements of design that can be applied across all touchpoints. And this has definitely been achieved. Good job, Strichpunkt!

Reduce and Recycle

The presence of flexible identities and living brands was to be expected at CXI_19 and did not disappoint. A simplicity of rules, reusable patterns and visuals, and a break with the separation between digital and print are leading principles of brand building and development in today’s world and in the foreseeable future.

The days of 200-page brand manuals and pixel-perfect media guidelines are numbered. Today’s brands can adapt, grow and survive. The brand is dead, long live the brand.

Conversational Experiences Meetup Review

Posted on 22 März 2019 by Paul Krizsan

The first of a new series of Meetups

We had a blast at our Conversational Experience Meetup in Berlin yesterday! Networking, meeting new people, gathering fresh and innovative ideas and perspectives.

While website-based online commerce and customer service have reached their limits, Conversational Commerce opens up entirely new, more human-centric opportunities for businesses. Getting ready for the conversational era has therefore become a priority for leading companies and brands worldwide. Chatbots and voice assistants continue to gain momentum.

Michael Bommer leads LivePerson’s European team as SVP EMEA and was the first speaker. With more than 25 years of experience, he advises some of the largest global enterprise companies on how to best leverage technology and AI to achieve a superior customer experience. As co-founder of relayr and Next Big Thing AG, Michael is a vocal part of Berlins start-up community, advocating for the opportunities of digital transformation in all industries.

Over the last 12 months, leading telcos began servicing customers through WhatsApp, enterprise brands have enabled conversations directly in their Google display ads, and retailers are selling products through Apple Business Chat and completing transactions through Apple Pay. Every aspect of the customer journey is now conversational. Michael made some great provocative and thought-enticing points on how call centres are a thing of the past and that the future of customer service is deeply intertwined with messaging.

Our very own Marco Spies then elaborated on the key aspects of what needs to be considered when designing a chatbot or voice assistant as they are increasingly part of the digital offering of brands. In particular he delved into the importance of a good match between bot personality and brand personality for the sake of differentiability, usability and brand recognition. To drive the point home, he spun an arch from the 80s TV series ‚Knight Rider‘ to Audis vision of their car voice assistant for the 2020s.

Ultimately Melanie Longdon of Liberty Global – the world’s largest international TV and broadband company – rolled out message-based customer support with bot to human handoff in early 2018 and told tales of satisfied customers and grateful support agents. For her the switch from call centres to messaging channels meant a tremendous leap ahead, which she underlined with impressive figures and statistics.

Thus, our short evening about conversational design became an evening of conversations on designing the future of conversing with customers in e-commerce situations!

For us the meetup was a resounding success and we absolutely intend on returning to the stage for future meetups, seminars and workshops, so keep your ears open!

An insight into the cosmos of voice assistants

Posted on 13 September 2018 by Paul Krizsan

In a world full of voice assistants, the choice is notoriously difficult. What are the strengths and weaknesses of the individual assistants and which ones are compatible with me? We have addressed these questions and created an overview of the most popular and competent voice assistants.

A brief overview

Google Assistant

Company: Google
Complexity: Android, iOS
Languages: 8
Topics: Travel, commute, search engine
Hotword: “Ok Google, Hey Google”
Features: Smart home integration, Android auto integration

Alexa

Company: Amazon
Complexity: Android, iOS
Languages: 4
Topics: Shopping, Entertainment
Hotword: “Alexa” (customizable)
Features: Smart Home integration, direct shopping on Amazon

Siri

Company: Apple
Complexity: iOS
Languages: 20+
Topics: Planning, productivity, every day life
Hotword: “Hey Siri”
Features: Apple CarPlay integration

Cortana

Company: Microsoft
Complexity: Android, iOS, Windows
Languages: 12
Topics: Planning, search engine
Hotword: “Hey Cortana”
Features: Integration to Alexa hardware

Our prediction for the future: Sprachassistenten und wohin führen uns diese mysteriösen Stimmen eigentlich?

Bixby

Company: Samsung
Complexity: Android
Languages: 2
Topics: Planning, Shopping
Hotword: “Hi Bixby” (customizable)
Features: Smartphone control, virtual shopping

Hound

Company: Soundhound
Complexity: Android, iOS
Languages: Englisch
Topics: Planning, search engine
Hotword: “Ok Hound”
Features: super fast, excellent at complex requests

Nomi

Company: Nio
Complexity: Car-Integration
Languages: Chinese
Topics: Vehicle management, daily companion
Hotword: unknown
Features: Very emotional through visualization, extension of the car

Personality and Voice

While features alone make up a small part of a good voice assistant, the personality of the virtual assistant is probably the most important part. The personality determines how committed we are to listening to the voice and how the hierarchy between human and machine is.

If the assistant is more humane, shows emotion and has as realistic a voice as possible, we tend to listen better and fulfill our wishes and requests through conversation. If the voice is tinny and clearly machine-like, we deal with the technology differently.

Our positioning of the personalities of the listed voice assistants (inspired of magenta.as)

The hierarchy between the user and the voice assistant can vary depending on personality: An always obeying, machine-sounding assistant is quickly regarded and treated as a will-less subordinate of the user, while a human-like personality tends to make us regard the assistant as an equal.

Due to the relevance of the personality, various voice assistants therefore also rely on visual supports, the so-called embodiment. Thereby, the voice assistant manifests itself not only in audible, but also in visual form, whether it is

We design and develop your voice assistant!

think moto develops brand-adequate conversational user interfaces for voice assistants and chatbots and also deals with questions of visualization and embodiment of voice assistants.

Voice assistants: So, where are these mysterious voices actually taking us?

Posted on 30 August 2018 by Paul Krizsan

Anyone currently walking past advertising posters for Google’s voice assistants is probably wondering where this voice assistant arms race will lead. Almost every year, the digital companions are equipped with new features and better capabilities. Will we ultimately be left with a homogeneous set of voice assistants that do everything perfectly, or will we face a wild flora of smaller assistants that are highly specialized but generally weaker, albeit in symbiotic relationships with each other? Or will it even look different?

Symbiosis

It is no secret that established voice assistants like Amazon’s Alexa and Microsoft’s Cortana have unique strengths, but also equally individual weaknesses. While Alexa demonstrates excellence among voice assistants in the areas of shopping, entertainment and as a companion outside the work environment, Cortana’s strengths lie in organizing daily routines and supporting the user’s productivity.

In May 2018, the manufacturers of the two assistants therefore announced a collaboration: In the future, one should not only be able to address and command Alexa via Amazon Echo, but also be able to call Cortana. Complete with Cortana’s voice.

This form of symbiosis is supposed to strengthen Alexa in particular, but it is also a sign that Cortana will probably not expand on the competitive stage. It is therefore more likely that Cortana will focus on deepening known topics.

Similar to Alexa and Cortana, the most popular voice assistants, Google Assistant and Siri, have weaknesses and strengths. The resulting gaps are now being filled by a new generation of voice assistants that can often only do a few things, but do them much better than general voice assistants such as those from Google, Apple and Co.

Companies like the U.S.-based Soundhound, whose Hound assistant shines especially when it comes to complex questions and commands, are either hoping to participate in the market alongside giants like Amazon by licensing their own framework. This allows corporations that would benefit from speech recognition and commands to use Soundhound’s technology without spending the resources to develop their own.

Voices and embodiment

While for manufacturers of mobile, smart devices in lieu of physical manifestation, it is primarily the voice that is the avatar of the personality, companies from industries such as smart home and automotive have the opportunity to visually lend a hand to the personality of the assistants. Whether physical or digital, this is referred to as embodiment, the lending of a visual language of form.

Amazon Echo

The embodiment can take different forms: Amazon can give Alexa coarser character traits through the design of the Echo products. Thus, the voice assistant does not appear feminine to the extreme, but rather neutral and open, educated and likeable.

Jibo

A good example of exaggerated embodiment is Jibo. Jibo is a curious and always joyful five-year-old in the cute body of a table lamp. By rotating the three body parts, the fun robot can, among other things, dance, tilt its head questioningly, and blink and show other emotions thanks to the eye in the display.

Although Jibo’s functions are limited and not nearly as elaborate as those of competitors, Jibo can convince with charm thanks to its physical form.

Nio Nomi

The automotive industry also sees a lot of potential in voice assistants. For many, our four-wheeled companions are already considered family members; you couldn’t ask for a simpler platform. Unlike smartwatches and smartphones, and not least because of the longevity and non-existent compulsion of portability, AI’s in cars can also take physical form. Much like Jibo, Chinese electric vehicle manufacturer Nio’s AI is intended to be perceived primarily as a social companion. Nomi – as Nio’s AI has been christened – can simulate and awaken in humans an astonishing array of human emotions thanks to a display above the car’s central console. It’s true that you feel like Luke Skywalker with a droid in a spaceship, but who can deny themselves those cute eyes?

A Forecast

However, the biggest technological leaps in the field of voice assistants are still happening with the market leaders. For example, in May 2018, Google unveiled a demo version of Google Assistant, which could independently make phone calls to humans with such authenticity that the people on the other end couldn’t make out the caller as an artificial intelligence. Google Duplex, as the demo version was called, based its humanity not only on the emulation of human speech, but also on the regular interspersion of filler words such as searching ahms and confirming mhhms.

Do you want to give users the feeling that they are talking to a real person or does the machine have to be recognizable as such?

That this involves the arbitrary deception of people and that the possibility for abuse is outrageously close is obvious. One of the biggest questions in the design of voice assistants in the coming years must therefore be the question of ethics: Do we want to give users the feeling that we are talking to a real human being, or must the machine be recognizable as such?

Dream vs. reality–A Beginner’s Guide to User Interfaces in XR

Posted on 13 August 2018 by Paul Krizsan

Five years after the arrival of VR technologies in our technology-centric society, extended realities are still a red-hot topic. The possibilities still seem limitless.

As an experienced digital brand agency, we are always expanding our in-house realities and can assure you that it is far from too late to jump in. Therefore, we would like to share with you some learnings as orientation, inspiration and good practices. Here are our golden rules for designing user interfaces for XR applications.

Learn from conventional media

As a pioneer in an extremely young and at least equally unexplored terrain, one naturally likes to rely on the familiar and one’s own experience from less immersive and more conventional media.

In Dead Space 3, the glowing spine indexes the damage taken, while the weapon displays the remaining ammo via hologram. Image source: dreamdawn.com

Game design is the ideal source of inspiration, as three-dimensional user interfaces have been explored there for decades. Since the days of the first consoles, game designers have been looking for ways to embed their user interfaces into the setting and narrative of their creations to maximize immersion.

While this is not true for all games, role-playing games in particular are based on the principle of immersing the user as much as possible in the virtual world.

Star Citizen not only presents the UI in a believable way, but also makes it visible to other players. Image source: vrnerds.de

Unbelievable interfaces create a disturbing dissonance here, which can quickly lead to the loss of the player’s attention. A central aspect that must also be avoided in XR applications in order to achieve the greatest possible immersion for the user.

Dirt Rally offers only authentic speedometer and cockpit displays for orientation. Image source: uploadvr.com

The solution is the diegetic UI, i.e. a user interface that exploits the spatiality of the application as well as the narrative and the setting in order to provide the user with credible information. This can be, for example, the clock on the protagonist’s arm, to which attention must be paid at regular intervals so as not to miss important events, or the dashboard in the cockpit of a vehicle, which can be used to read off speeds and other parameters.

More about UI in video games in this artikel of Anthony Stonehouse.

Choose the right interface concept

Probably the most important aspect for all design in XR is to always be aware of the aim and nature of the project. For example, a minimalist AR companion for concertgoers does not need to be absolutely immersive. Authenticity may not be the primary focus in such meta-interfaces. In this example, Diegetic UI is not superior to a regular, static interface, but only complicates the user experience unnecessarily.

Overall, user interfaces can be divided into three categories. Static UI, places information as a static overlay over the user’s field of view (this method is most popular with AR apps). In comparison, Spatial UI uses the three-dimensional environment to position the interface, such as a floating menu that curves around the user. Finally, there is Diegetic UI, based on which one does not break with the narrative of the application and the authenticity of the setting, and renders user-relevant information through believable shapes (a working clock on the wall, the display of a virtual smartphone, etc.). Before starting any project, it is therefore particularly relevant to weigh the pros and cons of the different platforms, as well as the different UI methods, and find the most effective solution for this project.

Design cross-platform and in patterns

The biggest challenge of Extended Realities – and also the biggest difference to conventional media – is the diversity of interaction design. The various disciplines could not be more different: While one application is controlled via motion tracking through gestures and hand movements, the next uses exclusively controllers and buttons, whereas another works only via gaze control.

This rich fauna of interaction possibilities requires strong design patterns and pattern libraries, especially for projects across multiple XR platforms, so that the user journey and user experience remain as consistent and user-friendly as possible.

Guide the user – gently and cautiously

Good UI: Clean, crisp and comprehensive. Image source: aixlab.com.

In the same vein, interface density is something to consider. Popups and crowded interfaces are already confusing on two-dimensional media and cause some headaches, but in XR – especially when you are positioned in the middle of the UI – chaotic and crowded UI has a claustrophobic effect.

Therefore, it is recommended to keep your distance first, let the user approach the virtual environment, and only then interact.

Final Thoughts

In any sense, one should not be put off by XR and its associated complexity. As a trailblazer, new aspects are explored every day and the opportunity to pioneer new niches is hardly so close in any other field. Thanks to open source software such as Unity and Blender, the entry threshold is extremely low and even laymen can build prototypes within a very short time. Go explore!

think moto goes rural–A Workshop-Experience

Posted on 30 Juli 2018 by Paul Krizsan

The day was here. My first time at think moto’s annual internal Branded Interactions Workshop. The expectations were high: I had already wanted to participate in a similar workshop half a year earlier as part of my studies before joining think moto for a semester. Unfortunately, nothing came of it then, but now the time had finally come.

For two days we were going to leave Berlin. Two days of workshop in the Mecklenburg Lake District with my head between nature and service design.

Friday, morning

Some of the colleagues have taken the workshop several times before, but there always seems to be a new component. It would never be boring, I was told. We pack the cars with plenty of provisions and workshop materials. I pick up conversations of Argentine barbecue and massive chunks of grilled meat. Did I mention that I love barbecue?

We arrive in the early morning. As a country bumpkin from Lower Saxony, I feel right at home in the green lowlands. The birds chirp pleasantly while we unpack, briefly enjoy the sun and set up the workshop.

Friday, noon

After a short snack, things get underway: Marco Spies, strategic partner at think moto and now head of the workshop, explains what the workshop will be about. The twelve participants are divided into three groups and the tasks are distributed. In order to oil our creative gears and gain experience with the Branded Interactions Design process, the task is to develop a new product or service for a specific brand. Of course, the thinking is user-centric, so each group is assigned a specific target group. When the starting signal is given, the three groups spread out across the site. The crackling of a freshly lit fire backdrops an intense brainstorming session.

For a short break, they head to the nearby lake. A colleague has brought fishing equipment and would like to treat us to some fresh catch. While he bobs on the lake with a small boat, the sun shines strongly on the surrounding forest. The scenery is picturesque, Bob Ross could not depict it more peacefully.

Friday, evening

After the short lunch break, the work continues, ideas are considered and discarded. Finally, the groups present their respective current status. Since our main focus is on creative methods, some presentations take on their own forms. Especially the lively role play of a group creates a cheerful atmosphere.

As dusk approaches, it is time to set up dinner. The rumors turned out to be true. Two massive pieces of meat cook on the two grills. In the meantime, vegetable chopping becomes a team-building task. Shish kebabs are created at breakneck speed and the work table would fit just as well in a factory hall of growing industrial nations.

In the late evening, all participants gather around the campfire. We sit relaxed around the fire, faces lit by twitching flames. Stories are told in the warming glow of the fire, while the topics take on an increasingly philosophical nature as time goes on.

Saturday, morning

Saturday morning we awaken gently to the aroma of coffee in the air and the songs of local birds. After a hearty breakfast and a team yoga session, it’s final spurt in the workshop. The groups meet one last time for final set up before it’s time to present the results of the last twenty-four hours.

Although there had already been presentations on interim statuses before, the focus was now for the first time on realizable products and services. All groups surprised with unexpected and very creative approaches.

Saturday, noon

With the official part of the workshop over, it was time to go to the lake once more to relax before most returned to the metropolis of Berlin.

It was the most intensive workshop I had the pleasure to participate in. Besides learning some new methods, I was also able to enjoy deeper insights into the Branded Interactions design process directly from the author. An experience that was definitely worth it.

Fare ye well, I’ll see you soon, forest.