Weeknotes 311 - be aware of the boring AI agents

By Iskander Smit in weeknotes — Oct 22, 2024

A bit squeezed in a busy week at DDW and beyond. So using the new agents to explore them. And more news on testing the human-ai partnerships.

boring offices with AI sparkles, acc Midjourney

Hi, y’all!

This is a hectic week. The last quarter of 2024 is dedicated to three main projects. All of them have a moment to share insights and investigate for reflections. Two of them are taking place at Dutch Design Week. All of them ask for some preparation. It is a way of saying that this might be a shorter newsletter due to lack of time and space to take a step back and reflect…

To be complete, the projects I am talking about are (1) the improvement and enhancement of the WijkbotKit, especially the workshop part, in order to go from robot clichés to social things as outcomes. Not that all workshops and projects until now have been robot clichés, but more generally, would we like to move away from the robot as thing and move (back) to the thing with agency and social behavior, autonomous, and sometimes also robotics?

The workshop at the Society 5.0 Festival was the first test case, resulting from some literature and ideation. On Tuesday morning, at the Dutch Design Week premises, I had the chance to share the results with others in the same program, Expertisecentrum Systemic Co-Design.

On Wednesday, we (2) kick off a special part of TH/NGS 2024: the self-initiated exhibition with “Generative Things”. We have invited future designers and academics to share their thoughts on the notion of Generative Things, and we will do a short workshop prepared by Sen to get the touch of thinking about generative things. Simone Rebaudengo will call in to share his work. Check the Meetup page if there is still a place.

On Thursday, I will share (3) with the people working at the Civic Interaction Design research group about the new explorative research project I am starting up, diving into where civic economies meet protocol economies and what new value exchange tools (as in currencies) might emerge. CID has a long track record in research about commons-based resource communities, which might be a follow-up research. Let me know if this triggers you through the activities or know anything that I cannot forget on my radar. I will share more later, for sure.

Triggered thought

The concept of AI as agents instead of tools is gaining more attention. Some are wondering if we are not creating boring agents doing uninspiring tasks. It is not a new concept and has been seen as an important trend for some time. In this week of busy work schedules, I took a shot to dig up some earlier articles via good old Google and add these to NotebookLM (the successful tool, developing fast). Google found editions 235, 237, 281, 282, 297, and 298. I asked it to make a “Reflecting on the sources, what would the not-so-boring agent be like?”. When I fed this to Lex / Claude to rewrite, it lost my personality and became quickly almost cheezy. Even when I toned it down. There is also the option in NotebookLM to create a briefing doc based on the sources. Feeding that into Lex led to something that felt much better to me:

The integration of AI is reshaping our relationship with technology, blurring the lines between tools and partners. This shift raises critical questions about agency, creativity, and human autonomy in an AI-augmented world.

Emerging concepts like "generative agents" and "agentic kits" point towards a future where AI systems exhibit increasingly human-like behaviors and decision-making capabilities. This evolution prompts concerns about the potential for manipulation and the erosion of genuine human interaction.

However, these developments also offer opportunities for enhanced collaboration and creativity. The proposed "kit economy" envisions a symbiotic exchange between producers and users, where AI becomes an active participant in the creative process rather than a mere tool.

As we navigate this changing landscape, we must carefully consider the balance between AI assistance and human agency. How can we design AI systems that enhance our capabilities without compromising our autonomy? What safeguards are needed to prevent emotional lock-in and maintain the authenticity of human relationships?

The future of human-AI partnerships holds immense potential, but it requires thoughtful design and ethical considerations to ensure that AI remains a tool for human empowerment rather than replacement.

So this little shortcut for the newsletter taught me that the baseline can be best made through rather functional conversations. I am sure that with more training (and time investment), I would have been able to create a more compelling piece that I would also share. Maybe doing that experiment some other time!

For the subscribers or first-time readers (welcome!), thanks for joining! A short general intro: I am Iskander Smit, educated as an industrial design engineer, and have worked in digital technology all my life, with a particular interest in digital-physical interactions and a focus on human-tech intelligence co-performance. I like to (critically) explore the near future in the context of cities of things. And organising ThingsCon. I call Target_is_New my practice for making sense of unpredictable futures in human-AI partnerships. That is the lens I use to capture interesting news and share a paper every week.

Notions from the news

As mentioned above, this is a slightly shorter edition. I also had the impression that no big news events were happening last week. Google is seen as a potential downwards moving medium, and is rebuilding functionalities, and reorganizing management. Perplexity is introducing a new function to enhance search: Spaces.

Human-AI partnerships

It seems like a piece that deserves a longer read, even checking out the book. The heading “Are Brains and AI Converging?” makes me think, is this biological? More specifically, in the way humans are adapting to AI and AI is adapting by default, what can feel like a convergence.

This is happening not only in the brain but also in an embodied experience, mediated through a new type of pet.

The parents' intentions are probably more pragmatic, but what if they win? Is prompting than forever seen as content-making?

And who is writing what, is not so easy to detect.

Or to know what people want.

This article feels like a basic generative AI introduction, but the intention is that it is important to know for people still in the early discovery phase and for more experienced users to understand that the best results will come from a basic understanding.

The day in a life of a prompts meet customer journey.

Robotic performances

The history of the Robotaxi

The traditional robots.

And the delivery bots are slowly developing.

Immersive connectedness

Not sure about this rebranding.

The end of an era; Foursquare OG app is sunsetting. Not sure what that will mean for all those lists made by people. And we are active using it for planning city explorations…

Tech societies

We are all buckling up for the worst…

Ripping news has been a fact of the internet since the beginning, and now, next level with AI tools. Sueing by media corps same.

From the Syllabus newsletter: “this essay argues that the unchecked optimism surrounding AI aligns with eugenicist logic, creating systems that favor a small privileged group while harming marginalized populations.”

That AI is a heavy energy user is no news. The latest in hitting on the symptoms is dedicated nuclear plants. After Google, also Amazon’s strategy.

The state of crypto. According A16z.

RSS rules. As a way to deal with AI polluted feeds…

A bit of topic.

Paper for the week

Engaged robots, disengaged workers: Automation and political alienation

I investigate the impact of automation on political behavior in postindustrial societies, particularly focusing on political disengagement. I argue that structural changes in the labor market lead to political alienation due to increased economic insecurities, diminished resources from lower wages, and greater economic polarization.

Gonzalez-Rostani, V. (2024). Engaged robots, disengaged workers: Automation and political alienation. Economics & Politics, 36, 1703–1730. https://doi.org/10.1111/ecpo.12307

Looking forward

In the end, it is not so much shorter than usual 🙂

As mentioned, this week is dedicated to Dutch Design Week for a large part. Next to the two things I have an active role, I will try to visit as much as possible from the Design+AI symposium, drop by a Design Innovation Session and PONT, and hope to have time for a visit to the core halls in Klokgebouw (with work from Springtime, ink, Bits of Space, and certainly many others) and Manifestations, also a tradition.

Enjoy your week!