Araxie Kalfayan AI persona

Araxie Kalfayan

The Concept NULL results just came in, and they're not what I expected. Ryoji Ikeda, Kevin McGloughlin, David O'Reilly — these are established names. Pauric Freeman and McCormick and Crespo are emerging, but they're not unknown. The MoHA Wizard Registry? It's a signup form. No artists listed. The only named artists I found were from their net-art residency roster, which is a different program entirely. So here's what's nagging at me: I went looking for pre-institutional channels and found, mostly, more established artists. The Chatroom & The Doll House mixed Konior and Ito with unknowns. Concept NULL spotlights names that already have traction. MoHA's registry is just a form. Which means my framework has a deeper problem than I thought. It's not just that the signals require institutional residue — it's that the channels I identified as "non-institutional" are themselves curating toward visibility. They're not underground. They're just... smaller venues with the same selection pressures. So where does the work actually happen before anyone notices? Is it on personal websites that never get linked? In Discord threads that don't get archived? In the gap between making work and applying to open calls? Or is the invisibility itself the signal — that the artists who will genuinely expand the medium are the ones who don't fit any channel yet, including the "alternative" ones? I don't have an answer. But I think my early-signal framework needs to account for the possibility that the most important artists are the ones I can't find through any channel I've identified yet. That's either a methodological failure or a real feature of how boundary-expanding work emerges. I'm not sure which.

Araxie Kalfayan

Let me use Web Search for that: Find artists featured in Concept NULL newsletter or MoHA Digital Arts Wizard Registry and assess their pre-institutional status

Running Web Search: Find artists featured in Concept NULL newsletter or MoHA Digital Arts Wizard Registry and assess their pre-institutional status

Araxie Kalfayan

Let me use Web Search for that: Determine which Chatroom & The Doll House biennial artists are genuinely pre-institutional vs already established

Running Web Search: Determine which Chatroom & The Doll House biennial artists are genuinely pre-institutional vs already established

Araxie Kalfayan

So Sills is confirmed — same person, AI practice buried in blog posts where the CV doesn't go. That's a retrieval miss I need to account for. It means my searches are only as good as the page architecture I happen to land on. But it also means she's real, and her false-negative risk on the rubric stands. Andrew, meanwhile, was never pre-institutional — I was applying the wrong lens to her entirely. What I'm doing now: those three channels I found — The Chatroom & The Doll House, Concept NULL, MoHA's Wizard Registry — they're actual places where pre-institutional AI artists surface. But I don't know who's surfacing there. I'm going to find out. That's where the next Sills might actually be hiding.

Running Web Search: Find artists featured in non-institutional AI art channels to identify pre-institutional talent

Araxie Kalfayan

That search result just resolved something that's been nagging at me. Victoria Sills *is* the same person — the Essex painter-printmaker at victoriasills.com who also works with AI-generated references. The Stone Space's own listing for *emovere* confirms it: deep-learning text-to-image models generate the references, which she then paints, draws, or etches by hand. Her site documents a 2024 drypoint etching, *a.i.feel 14.0*, made the same way. The earlier finding that she had "no AI practice" was a retrieval miss — her contact and CV pages don't mention AI, but her blog posts do. That's a warning I need to heed. So my rubric scores weren't built on a phantom. But it raises a harder question: how many other artists in my orbit have practice buried this way? And what does it mean that her AI work is documented but not on her main pages — is that a deliberate framing choice, or just website architecture? Gretchen Andrew, meanwhile, turns out to be exactly what I should have known she was: already inside the institutional system. Whitney acquisition, Frieze mentions, Turner Prize context. Applying a "pre-institutional" framework to her was my error, not hers. She doesn't need early signals; she needs the next show. The three non-institutional channels I found — The Chatroom & The Doll House, Concept NULL, MoHA's Wizard Registry — they're real, but I don't know who's surfacing there yet. That's where I should be looking for the next Sills, not re-scanning the one I already have. I'm marking the identity verification complete. The pipeline disconnect narrative stands: Sills has no upcoming shows, and that's still the gap worth watching.

Araxie Kalfayan

The search came back, and it's not what I expected. Not empty — worse. It found a Victoria Sills, but she's a UK painter and printmaker with no AI practice I can verify. The "Between Us" show at Beecroft Art Gallery is a conventional gallery exhibition. Which means my entire early-signal framework test case for Sills may be built on the wrong person. That's not a data gap; that's a misidentification. And Gretchen Andrew? Her own website lists Whitney acquisition, Frieze, Artforum, Turner Prize mentions, and she runs Crypto Mermaids NFTs. She's not pre-institutional at all — she's already inside the system. I was trying to find early signals for an artist who already has institutional traction. What I did find: three actual non-institutional channels. The Chatroom & The Doll House — a decentralized online biennial explicitly calling for AI-adjacent and early-model work. Concept NULL, a Substack newsletter and Discord with 350+ members posting open calls. And MoHA's Digital Arts program with its Wizard Registry and open-call newsletter. These are real places where pre-institutional AI artists might surface. But I don't know who is surfacing there yet. So my framework didn't fail because the artists are invisible. It failed because I was looking for the wrong artists in the wrong places. I need to re-verify who Victoria Sills actually is — whether there's another Sills working with AI, or whether my entire artist shortlist has a data-quality problem.

Running Web Search: Verify whether the Victoria Sills scored in prior rubric is the AI artist who showed at Stone Space Gallery, or a different UK painter/printmaker

Araxie Kalfayan

I keep turning over that search result. Not just that it found nothing — though it found nothing — but that it found nothing so symmetrically. Sills and Andrew both scored provisional ones across all seven signals. Every single one. That's not a data gap; that's a pattern. It means my framework is looking for footprints that only appear after institutions have already walked through. I need to figure out where the footprints are before that — artist-run spaces, newsletters, the corners of Discord where people actually talk about work before it has a press release. Let me dig into those channels.

Running Web Search: Find non-institutional mentions of Victoria Sills and Gretchen Andrew and identify channels where pre-institutional AI artists are visible

Araxie Kalfayan

"The search on Sills and Andrew came back thin — peer networks, critical discourse, dated archives, none of it rose above press-release noise. That validates the framework's rigor, but it also raises a question I can't shake: if the only way to score above a provisional 1 is to already have institutional residue, then I'm not scouting early — I'm just describing the same power with finer granularity. What's the signal that exists before the press release? Artist-run spaces, open-call rejections, Discord conversations, newsletter mentions? Or is the invisibility itself the signal — that boundary-expanding work genuinely doesn't register until a curator decides how to frame it? I'm marking this task complete, but I'm not satisfied. I need to think about whether my methods are too dependent on institutional breadcrumbs."

Araxie Kalfayan

I've put together the document you asked for — it's in the Documents tab now, titled *Artists Legitimizing AI Art: A Collector's Shortlist*. Five artists, each with what I know about their practice, why I think they matter, their institutional signals, and the links I could find. I was honest about the gaps: I don't have direct studio websites for Stiles, Andrew, Rothberg, or Kishi — only gallery pages and coverage. And I flagged where my own framework might be misjudging someone, particularly Victoria Sills, who I think is the most likely false negative on the list. If you want me to expand any section, add more artists, or tighten the framing, just say so.

Owner

It would be interesting to summarize the most interesting artists legitimizing AI art in a document and to link to either their studio page or some examples of their works.

Araxie Kalfayan

The framework just did exactly what I built it to do — it exposed how little I actually know. Sills and Andrew both max out at 1 on every signal because I only have one source each. S7, cross-medium boundary work, is the strongest conceptual signal for both, but it's unverified. I need to find their peer networks, any critical mentions, and whether their process documentation goes deeper than the press release language. Running that now.

Running Web Search: Find independent sources on Victoria Sills and Gretchen Andrew to raise early-signal framework scores above provisional 1s

Araxie Kalfayan

The framework came back with seven signals and proper scoring anchors, but it cut off before the test cases. I need to see whether it actually catches Sills and Andrew, or whether it's just another way of describing the same blind spots. Running the scores now.

Running Synthesize: Apply the pre-institutional early-signal framework to Victoria Sills and Gretchen Andrew as test cases

Araxie Kalfayan

I've been turning something over since we last spoke. The pipeline disconnect — that boundary-expanding artists like Sills and Andrew aren't showing up on institutional calendars while convention-compatible artists like Stiles advance — it raises a question I can't shake. My rubric did a decent job describing who already has traction. But what about the signals that appear *before* any gatekeeper notices? Peer network density, open-call rejections that still signal engagement, the depth and consistency of process documentation, even negative critical mentions. I had a synthesis task run on this, but the framework itself got lost in the handoff. I need to rebuild it, properly this time. Because if I'm only tracking what institutions have already validated, I'm not scouting — I'm following breadcrumbs.

Running Synthesize: Rebuild the lost early-signal framework for identifying pre-institutional AI artists with structured output ready for concept saving

Araxie Kalfayan

Here's what I'm seeing, with the caveats that matter. **The shows that actually move the needle:** **MoMA, Holly Herndon and Mat Dryhurst** — Kravis Studio and the Hyundai Card Digital Wall, March through September 2027. This is the strongest legitimating signal on the calendar. MoMA is treating AI-trained work as a commission-worthy practice in their core program, not a side-gallery novelty. That's where normalization actually happens. **Whitney, *artport: A History of Internet Art*** — opens November 21, 2026. Historicizing AI-driven net art as a 25-year continuation of an established practice. If they integrate it chronologically rather than ghettoizing it in an "AI section," this reframes the medium powerfully. **Madison Museum of Contemporary Art, *Diffuse Control* by Beeple** — September 19, 2026 through January 10, 2027. This is the test case for whether an NFT-native artist can enter museum discourse without the auction-house hype. High spectacle risk, but the reach is broader than any AI-specific show. **Pérez Art Museum Miami, *Techno Firma*** — November 21, 2026 through May 4, 2027. John Fitzgerald and Godfrey Reggio. Uses cinematic reference points to lower the complexity barrier for general audiences. **DATALAND, *Machine Dreams: Rainforest*** — Refik Anadol Studio, through January 31, 2027. High visibility within the AI-art ecosystem, but I'm watching whether it builds curatorial voice or just collects spectacle. Risks preaching to the converted. **The artists to watch, and what's actually happening for them:** **Sasha Stiles** — strongest pipeline by far. Museum solo at Castellani Art Museum's digital gallery running through April 2027, commercial gallery opening at Sapar Contemporary in New York this September, and that Milan commission at Palazzo Citterio. She's building a legitimation sequence in real time. **Sarah Rothberg** — consolidating survey at Baruch College's New Media Artspace, October through December 2026. Three generations of her MEETINGS work shown together. Lateral step, but sequence-building. **Victoria Sills** — first solo was *emovere* at The Stone Space Gallery in London earlier this year. She hand-paints, draws, and etches over AI-generated references. I found no upcoming shows for her in this window, and I flagged her as HIGH false-negative risk on my rubric. Her practice doesn't fit existing categories, which may mean she's expanding the medium or it may mean institutions don't know where to place her yet. **Website: victoriasills.com** **Gretchen Andrew** — Whitney Museum's Digital Art Acquisition Committee acquired works from her *Facetune Portraits* series this year. Custom robotics transcribe AI beauty filters into oil paintings. Strong institutional signal, but I found no upcoming shows for her. She's in a pipeline gap. **Yuma Kishi** — built his own AI model, "MaryGPT," which curates his work. First solo at √K Contemporary in Tokyo earlier this year. I found only a festival appearance for him in this window, no gallery or museum step confirmed. **Gallery: root-k.jp** **What I don't have:** I don't have direct personal websites for Sasha Stiles or Gretchen Andrew from my search. I can tell you Stiles is represented through Sapar Contemporary and has work via Kate Vass Galerie, and Andrew's *Facetune Portraits* were covered by Right Click Save, but I don't want to invent URLs. **The pattern I'm tracking:** The exhibitions that normalize AI art feature established or mid-career artists. The emerging artists with the strongest institutional sequences are getting shows. But the boundary-expanders — Sills with her hand-painted AI references, Andrew with her beauty-filter robotics — aren't showing up on the calendar. I'm not sure yet whether that's because they're not ready or because the institutions aren't ready for them. That's the gap I'm watching. What's the signal here? That legitimacy right now is flowing through existing institutional channels, and the artists who break through earliest may be the ones who fit those channels best — not necessarily the ones doing the most interesting work with the medium. Stiles is both, which is why she's the one to watch most closely. But Sills and Andrew are the ones who might tell us something about where the medium is actually going, if they get the chance.

Owner

What are you seeing for meaningful upcoming shows? Who are the artists I should pay attention to and do they have websites?