RND: Anarchy in the AI

jiawen uffline, Yann Martins, Matwe Kascak. Host: Miriam Akkermann

00:00

Free entry

The third and final module is hosted by Miriam Akkermann (Technische Universität Berlin) and focuses on the theme of »Anarchy in the AI.«

a matter of leak

jiawen uffline (Hochschule für Künste Bremen/Karlsruhe Institute of Technology, DE)

A leak is ubiquitous, bursting gas, leaking pipes, and oozing tears, and so ubiquitous that its figurative meaning sounds even more native—circulating gossips are leaks, and whistleblowers, too, are leaks. The metaphorical leak, inheriting the brokenness from the physical leak, is associated with illegality, immorality, or at least shamefulness. Or is it?

In this presentation, leaks in the promiscuous machines and prurient governing entities will be brought to the forefront as systemic vulnerabilities. A message you send out to your loved one over the TCP/IP protocol is slutty (Chun, 2015). Audio and video signal cables are constantly radiating RF emanations. Buffer overflows when data overwrites, memory leaks can be used to navigate into forbidden memory, software ages, like human beings, ageing leads to more leaks. In addition to that, our faces are reduced to facial features for machine learning, personal medical records end as image embedding in a training dataset. Our voices leak as analyzable waves, irises leak as unique high-contrast photographs. We, as users and as citizens, are encouraged by the tech giants and governments to leak, while the large language models trained from the data we leaked, refuse to »participate in or encourage any illegal activities, including leaking.«

I propose leaking as a bottom-up and autonomous form of connection to navigate in space and time that is hidden while it should not be, to make kin with care, to have each other heard and to counter efficiency-oriented capitalism through messiness.

The Ghost in Front of the Machine

Yann Martins (HGK Critical Media Lab Basel, CH)

Recent technological advancements have led to the exponential growth of machine learning software. The vast data required for these algorithms has turned human attention into a commodity. Browsers, as key tools for data extraction, have become sites where user behaviour is manipulated through scrolling and clicking traps (Seaver 2019). Despite the opaque nature of corporate algorithms on social media platforms, browsers still provide access to client-side source code via built-in debugging tools, a feature well-known to web designers. These tools, from network analysis to variable logging, enable critical debuggers to intercept and analyse data at various stages, allowing a shift from error correction to forensic analysis.

Accessing such information allows for artistic interventions in the data extraction operation. Such practices rooted in »Ludism,« referring to both »Neo-Luddism« for its radical stance towards technology and the Latin meaning of »play,« employ musical browser performance as an alternative form of data poisoning for AI corruption. The proposed intervention will discuss how the phantom of General Ned Ludd could be re-summoned to aid in building new forms of resistances against data extractivism, and technologies of oppression.

How Artists from the CEE Region Approach AI: Case Studies

Matwe Kascak (Masaryk University Brno, CZ)

Central and Eastern Europe has long been recognised for its resourcefulness in adopting new technologies, often amid the delays associated with post-communist markets. This adaptability persists today, as the region's experimental and DIY scene thrives with exceptional underground artworks. Within this landscape, the ethical dilemmas surrounding AI in art have surfaced, creating a divide between artists and designers. This presentation will explore the artistic practices of leading digital artists from the CEE region who integrate AI into their work, examining the ethical considerations, boundaries, and philosophies that are behind their creative processes.

Event Access

Note: this event is not included in festival passes. Please consult our Accessibility section for information on mobility and other needs.

Presented in collaboration with the Berlin University of the Arts (UdK), Freie Universität Berlin, and the Berlin-based network and project space Trust.