Tootfinder

Sentenza di Cassazione 22791/26 sui termini sanzionatori del Garante – “Un’altra vittoria così e sarò rovinato”
https://www.istitutoitalianoprivacy.it/202…

Nachfolger von Windows: Staatskanzlei Kiel stellt auf Linux um
Schleswig-Holstein treibt den Wechsel von Microsoft auf Open Source voran. Nach E-Mail und Office läuft die Kieler Staatskanzlei nun weitgehend unter Linux.

Nachfolger von Windows: Staatskanzlei Kiel stellt auf Linux um
Schleswig-Holstein treibt den Wechsel von Microsoft auf Open Source voran. Nach E-Mail und Office läuft die Kieler Staatskanzlei nun weitgehend unter Linux.

Hong Kong-based Qupital, which offers cross-border ecommerce financing to SMEs, raised a $300M Series C led by M Capital as it weighs a possible IPO (FinTech Global)
https://fintech.global/2026/09/15/qupital-eyes-ipo-after-landmark-300m-series-c-round/

Qupital eyes IPO after landmark $300m Series C round
Qupital, an AI-driven trade finance platform focused on cross-border e-commerce, has raised $300m in combined new capital, anchored by a Series C funding round aimed at accelerating its expansion into overseas markets. The financing was led by Asia-based asset manager M Capital, with additional asset-backed securities commitments from Mitsubishi UFJ Financial Group (MUFG) and Quester […]

Orientation Reading by Production Vision-Language Models on Optotype Charts: A Controlled Multi-Model Evaluation Across Reasoning Modes, Prompts, and Access Modalities
Shahryar Wasif, Avneek Sandhu, Bin Hu
https://arxiv.org/abs/2607.16595 https://arxiv.org/pdf/2607.16595 https://arxiv.org/html/2607.16595
arXiv:2607.16595v1 Announce Type: new
Abstract: OBJECTIVES: Vision-language models are increasingly used to interpret medical and everyday images through consumer chat interfaces, yet their ability to read orientation - the single perceptual operation tested by the tumbling-E acuity optotype - is poorly characterized on the surfaces through which they are actually used. METHODS: We evaluated four production vision-language models (referred to as Claude, GPT, GROK, and Gemini) through their consumer chat interfaces on a locked set of seven optotype charts: four uniform tumbling-E charts (one per cardinal orientation), two mixed-orientation tumbling-E charts, and one Snellen letter chart as a specificity control. Each model was run in two reasoning modes (Fast and Thinking) under two prompt variants (with and without an explicit orientation-decoding rule) by up to three operators. The corpus comprised 920 scoreable trials and 50,420 glyph judgements. The primary outcome was glyph-level accuracy against the chart's designed orientation, summarized with Wilson 95% confidence intervals. RESULTS: Accuracy ranged from 43.0% to 97.0% across models on identical charts, and the strongest model depended on reasoning mode (GPT 97.0% in Fast mode; GROK 96.6% in Thinking mode). Errors were not random but collapsed onto a model-specific attractor direction. Models were 96-100% internally self-consistent yet ranged widely in accuracy, dissociating reliability from validity. An answer-key-free ensemble-consensus estimate tracked accuracy closely (r = 0.998). For one model, consumer-interface accuracy fell 25-27 points below programmatic access, almost entirely on a single orientation. CONCLUSIONS: A single accuracy figure conceals clinically relevant, orientation-specific failure modes; vision-language models should be evaluated along multiple axes and on the deployment surface before image-interpretation outputs are trusted.
toXiv_bot_toot

I'm looking for something that I absolutely *could* DIY but am pretty sure exists as a commercial product.
* Large format red/black/white (yellow optional) eink display, 7 inch or so
* Ethernet or wifi
* Local push based update (i.e. I can make a POST request or something directly to the device to dump an image onto it without needing to go through a cloud server or be on a VLAN with internet access. Timer-based pull update (device downloads an image at regular intervals)…

At the end of the month and early October, I’ll be present at three major Open Source conferences:
- X Developer Conference (XDC 2026) in Toronto
- Linux Plumbers Conference (LPC 2026) in Prague
- (Embedded) Open Source Summit Europe (E-OSS-E/ELC-E 2026) in Prague
Over the 3 conferences, I’ll speak about how we should modernise the Linux display panel support for three different audiences: Graphics Developers, Linux Maintainers and finally Linux Developers from the indus…

Little tip when linking to source code in your docs, etc.
You can link directly to lines using anchors if the web view of your source control system supports it. e.g., for Codeberg:
https://codeberg.org/small-tech/jsdb/src/branch/main/lib/QuerySanitiser.…

jsdb/lib/QuerySanitiser.js at main
jsdb - A zero-dependency, transparent, in-memory, streaming write-on-update JavaScript database for the Small Web that persists to a JavaScript transaction log.

Source: OpenAI's VP of sales in the Americas, Kaylin Voss, resigned a week after Denise Dresser left, prompting others on the sales team to consider resigning (Laura Bratton/The Information)
https://www.theinformation.com/briefings/exclus…

Exclusive: Key OpenAI Sales Executive Kaylin Voss Resigns
Kaylin Voss, OpenAI’s vice president of sales in the Americas, has resigned a week after her former boss, Denise Dresser, left the company, according to a person familiar with the matter. Voss had been at OpenAI for just five months. Dresser, who was chief revenue officer, had left after nine months at OpenAI. The company hired Dali Rajic from Wiz to replace her. Like Dresser, Voss had come

Mehr als 50 Millionen Euro für E-Auto-Prämien - Tesla vorne
Seit knapp zwei Monaten gibt es wieder eine staatliche Förderung für E-Autos. Wie viel Geld schon floss und welche Marken profitierten.
https://www.

Mehr als 50 Millionen Euro für E-Auto-Prämien - Tesla vorne
Seit knapp zwei Monaten gibt es wieder eine staatliche Förderung für E-Autos. Wie viel Geld schon floss und welche Marken profitierten.

Anthropic is partnering with Accenture to embed evaluators within Anthropic, including red teaming models and conducting alignment assessments (Anthropic)
https://www.anthropic.com/news/accenture-embedded-evaluation

Partnering with Accenture on embedded evaluation
We’re partnering with Accenture on independent evaluation of frontier AI—part of our recent commitment to embed evaluators at Anthropic. Both we and Accenture expect to invest at least $1 billion to build capacity in this area over the next five years.