tristanj

5 ms·

61.
by tristanj 1mo ago
There's some glaring mistakes in your framing. First, this is an unnamed OpenAI employee speaking, not OpenAI the organization. Second, you miscomprehended the article. The employee did not "threaten to ruin a prominent researcher
62.
by tristanj 1mo ago
The rumors that Anthropic had solved a millennium problem were absolutely everywhere last week. I'm not surprised at all that OAI took their own stab at it.
63.
by tristanj 1mo ago
Because very few people actually have access to these logs, all access is monitored and recorded, and improper access will get you fired. It's not worth risking your job over something like this.
64.
by tristanj 1mo ago
That's a stretch. The Luis and Diego paper was published in 2023 and is included in every frontier model's training dataset. An AI model could independently choose the same path route as Luis and Diego, without access to Buckmaste
65.
by tristanj 1mo ago
This LAN house was featured in a LTT video. There's an entire house walkthrough https://www.youtube.com/watch?v=97Y0MVUgjOw
66.
by tristanj 1mo ago
Anthropic is currently in an IPO quiet period and cannot publish such information.
67.
by tristanj 1mo ago
Chinese models are trained on dubiously collected model traces from Claude/OpenAI that are purchased from model routers. All the major Chinese models use this data. That's one reason they've been able to catch up with Anthrop
68.
by tristanj 1mo ago
This conclusion is flawed. It's unclear at this point if OpenAI's model or employees actually looked at or stole the author's data. Having worked at large companies before, I'm leaning towards no, since very few employee
69.
by tristanj 1mo ago
Yeah these are major accusations. But the story is incomplete, the conversation is missing a lot of details. It's not clear who was working on what, and when. The entire thing feels rushed, like they wanted to get this result published
70.
by tristanj 1mo ago
Full mastodon post: https://mathstodon.xyz/@tristanbuckmaster@mastodon.social/11... Mathematical explanation by Terrance Tao: https://mathstodon.xyz/@tao/117233527638291447 It seems there is much
71.
by tristanj 1mo ago
I don't understand the value proposition here. This device is selling for $1,999. In the product page on their website, they show the device plugged into a MacBook Pro. A MacBook Pro already has excellent ability to run local LLMs, so
72.
by tristanj 1mo ago
This article is very shortsighted. If you think Meta is bad now, just wait until Zuck is replaced by a soulless, profit-obsessed corporate CEO whose primary job is maximizing shareholder profit. Meta as a company can get far worse. People
73.
by tristanj 1mo ago
That's hilarious, considering that by Gary's previous definition of AGI (the one he conveniently forgot about and never brings up again), we already have AGI. https://news.ycombinator.com/item?id=49207797
74.
by tristanj 1mo ago
OpenAI was in serious financial difficulty in 2024/2025 but that's no longer true in 2026. Their revenue blew up massively in the past couple months and their ARR passed $40B/yr. Most people haven't updated their priors
75.
by tristanj 1mo ago
This is an invalid comparison. There were enormous discounts on the Prologue, up to $20,000 off. https://electrek.co/2025/09/23/honda-offering-over-20000-off...
76.
by tristanj 1mo ago
Edit: Misread
77.
by tristanj 1mo ago
French Guiana is not a colony, it is a region of France. People who live there have the same constitutional rights as anyone living in mainland France. It has the same status in France as Alaska does in the US.
78.
by tristanj 1mo ago
You're reading way to much into this. Physical gold bars are transferred via plane between London and New York every day. The gold flies in the cargo holds of regularly scheduled commercial passenger flights. The quantity transferred d
79.
by tristanj 1mo ago
Moving the goalposts to a generic U.S.-decline narrative does not answer the question: How does Taiwan "switch allies" to the adversary that's trying to annex them?
80.
by tristanj 1mo ago
Switch allies, to whom? The PRC? Do you realize PRC (Mainland China) and Taiwan are in a civil war? Taiwan switching allies to the PRC literally means admitting defeat. Recall that Taiwan only exists because during the Chinese civil war, th
81.
by tristanj 1mo ago
Why do people comment when they understand so little about the China/Taiwan issue? It has never been about semiconductors, that topic only exists as a post-rationalization of why the U.S. should continue to protect Taiwan against PRC
82.
by tristanj 2mo ago
Source: https://x.com/mweinbach/status/2095596867081916653
83.
macOS 27 Simulator by OpenAI Astra(macos-27-simulator.mweinbach.chatgpt.site)
2 points by tristanj 2mo ago|2 comments
84.
by tristanj 2mo ago
Something similar happened to Broadway in New York City. It used to be a main thoroughfare, packed with cars in bumper-to-bumper traffic. But after the city decided to pedestrianize it, by banning cars, adding bike lanes, and converting muc
85.
by tristanj 2mo ago
The video shows dozens of seats, so they are not only aiming for short haul cargo.
86.
by tristanj 2mo ago
You've missed the point. The engineers went too far optimizing for weight, to the point of sacrificing usability.
87.
by tristanj 2mo ago
The interior of the plane is clearly designed by an engineer. They stripped everything down to save weight, passenger comfort be dammed. The seats don't even look comfortable. They need to hire an actual designer, if the plane weighs 0
88.
by tristanj 2mo ago
"gpt-6-astra" has been staged on the OpenAI API https://x.com/synthwavedd/status/2095184148981842161 The OpenAI Responses API now returns a 404 Not Found for "gpt-6-astra", where garbage&#x2
89.
by tristanj 2mo ago
If it's not clear what's happened: The launch was scheduled for 11am Pacific time. The press embargo broke at 11am, and we saw a flurry of press articles by Axios, TechCrunch et al. The model has appeared on the ChatGPT API. But t
90.
by tristanj 2mo ago
GPT 6 Astra benchmarks https://cdn.thenewstack.io/media/2026/09/358eb84a-screenshot... Performance is significantly higher than Fable 5.1 Source: https://thenewstack.io/openai-gpt6-astra-bench

More ›