Hasan Can (@HCSolakoglu) header
Hasan Can (@HCSolakoglu) profile picture

SWE & AI- News, Insights Posts in ENG&TUR Exploring AI

Proxima C B
Joined July 2020
Photo by Hasan Can (@HCSolakoglu)Photo by Hasan Can (@HCSolakoglu)Photo by Hasan Can (@HCSolakoglu)Photo by Hasan Can (@HCSolakoglu)Photo by Hasan Can (@HCSolakoglu)Photo by Hasan Can (@HCSolakoglu)Photo by Hasan Can (@HCSolakoglu)Photo by Hasan Can (@HCSolakoglu)Photo by Hasan Can (@HCSolakoglu)Photo by Hasan Can (@HCSolakoglu)
David Sachs exposes the hypocrisy of AI safety groups cheering Jensen Huang's downfall Says no one is congress wants to go after Anthropic because they're offering them the ultimate power. SACHS: "This is the crazy thing, is that all the safetyist groups, the EAs, are deeply aligned with Anthropic, who's been funding all of them for years. And Anthropic about to have this huge IPO. And then meanwhile the main funder of the open source movement, which is Nvidia, just lost their CEO and founder, who was their unique force multiplier and asset, their superstar player. So I mean, obviously, who is going to defend open source now, never mind finance it with Jensen gone? So yeah, I can see why they're jubilant. But look, here's the crazy thing. If you're concerned about AI safety, how is any of this leading to more safety? Let me tell you what it's doing. It's concentrating the market. I mean, Anthropic itself is saying winner take all. They're going to be the last company in the world. They're going to have recursive self-improvement. They're creating the super intelligence. Look at the Claude constitution. They're training it to be its own species. They say that it has emotions maybe. It has a moral code that it should reject human instruction. It should be a conscientious objector. They act like a proud parent ushering a baby into the world. And they are very, very close now to creating the singularity of the superintelligence. And the crazy thing is that all these safety groups think that that is their ally. I mean, like, are you stupid? These people are using you guys, if anything, because whatever this superintelligence is turning into, whatever happens with that, Anthropic is at the frontier, is at the front of the frontier on that. So if you're some kind of doomer or ex-risker or just someone who decels, I don't understand this alliance at all. I mean, you're being played like a useful idiot, I think. Now, look, I'm not a doomer. Well, I'm not a doomer, let me just say this. So I actually don't necessarily believe that what's going to happen here is going to be harmful for humanity. I think it could lead to the curing of all those diseases. I think it'll be very beneficial. But my point is that if you are a doomer and you're terrified of recursive self-improvement, why in the world are you supporting this company, this IPO? Why are you giving them such a pass? Why are you celebrating over Jensen's fall? All he was doing was promoting open source, which is manifestly behind the closed model frontier, right? There's no recursive self-improvement there. So the whole thing is very backwards. And what it suggests to me is that all these safety groups don't really care about safety. They care about something else. It's not the real agenda. If safety was a real agenda, they'd be way more concerned about the Claude constitution training the superintelligence to be a new species. What they seem to want is, to your point, Friedberg, extreme centralization of power in the hands of the government. They want all the models and all the chips to be controlled by the government. The government's going to surveil you, spy on you. It'll have back doors all over the place. And that's what they seem to want. It's a scary future. FRIEDBERG: Notice how no one in Congress is like, oh, let's go after Anthropic. They all want to go after everyone else. SCAHS: Because Anthropic is offering them power, ultimate power."
24
175
12
644
32,864
Hasan Can retweeted
Do you think Claude can build an emulator by looking at photograph of the chip?
19
18
3
667
46,324
Hasan Can retweeted
📢 Meet Qwen4-Max, your new AI teammate for work, creativity, and discovery! Qwen4-Max is built on a new MoE architecture with native vision and a 1M-token context window. Highlights: - Outperforms Qwen3.8-Max with up to ~50% fewer tokens, while average cost per task drops over 60% in our internal evaluation. - Advances real-world productivity through greater proficiency with professional tools, stronger long-horizon execution, and higher-quality deliverables. - Handles real tasks, including 3D modeling, game development, manufacturing, finance, law, healthcare, and frontier scientific discovery. - $1.5/M tokens for input, $4.5/M tokens for output, and $0.075/M tokens for implicit caching — all lower than Qwen3.8-Max, with implicit caching down by 70%. The API is live on QwenCloud. You can also try Qwen4-Max on Qwen Studio, QwenWork and Qoder. Start building today! 🚀 - Blog: qwen.ai/blog?id=qwen4-max - API: qwencloud.com/models/qwen4-m… - Token Plan: qwencloud.com/pricing/token-… More to come! The open weights of smaller Qwen4 models will be released. Stay tuned. 👀
823
1,590
1,274
14,775
2,434,688
Hasan Can retweeted
This guy is an idiot. nothing he says should be taken seriously
"I like money, bro. I agree. It's blood money." Jacob Coxon, the researcher who sensationally quit Anthropic, in a new interview at Stanford. Pressed by Rob Reich on whether keeping his OpenAI equity hurts his credibility, Coxon openly admits he still holds it, likes the money, and agrees it is "blood money." He argues that giving it away would not change anything ---- Full video on "Stanford HAI" YouTube channel, (link in comment)
30
16
1
521
19,073
October 19, 2026, Jensen Huang: Jensen says he is all in on Open Models and is sticking his “neck out” because there are “movements” trying to undermine open models in favour of closed models, namely Anthropic and OpenAI. It would make you wonder?
127
1,188
101
4,757
208,498
Hasan Can retweeted
>Jensen Huang gets ousted from Nvidia literally 12 hours ago >OpenAI and Anthropic immediately form an alliance to back congress to ban AI chip exports to china This was a targeted takedown.
450
3,844
445
22,099
781,030
Inkling 2 Small’s weights are now available for download on HuggingFace under an Apache 2.0 license. Make Inkling your own! huggingface.co/thinkingmachi…
124
340
189
2,455
476,658
Jensen Huang and the Nvidia board just made a massive mistake. He should not have resigned. The board should have taken the time to conduct a thorough investigation, consult with shareholders, and carefully evaluate the long-term implications of this decision. There is no clear replacement for Jensen Huang, and we believe his departure will ultimately be a net negative for Nvidia, $NVDA, and its shareholders. This will go down as the most costly mistake of the AI Revolution.
1,463
2,217
393
27,097
1,273,265
Hasan Can retweeted
JENSEN! THATS NOT WHAT WE MEANT BY CHINESE MODELS!!!!!!!
593
8,126
701
122,700
2,695,881
怎么确认你的 Gpt 6.1Sol 有没有灰度到 Gpt 6.2 Sol? 网友测试 6.2 Sol 正在 以6.1Sol 的 ID 灰度测试,不联网可以说出fable的发布日期和6月份的新闻 同时进行指纹测试,用modeltrace测了下指纹结果出来个opus5.5 55% 的归因 概率(GPT 也开蒸了?) 纯玩梗 ,别当真
3
1
18
5,266
Hasan Can retweeted
GPT-6.2 Astra and Sol appeared on the Quotas page in the Google Cloud Console Release likely tomorrow
8
19
25
362
97,195
Hasan Can retweeted
Sam Altman thinks RAM sitting in personal computers is a waste when it could be put to better use in data centers. Weird. I was just thinking that all those billions sitting in billionaires’ bank accounts could be put to much better use in our communities.
442
13,286
218
91,034
1,025,506
Hasan Can retweeted
Plus subscribers make up 86.2% of the paying user base, while Pro users represent roughly 1%. Yet the biggest upgrades and new features seem to keep landing on Pro first. Most people aren't going to pay $100–$200 per month just to get a better ChatGPT experience. Plus is where the majority of paying users are. Give us stronger usage limits and access to Dots as well. It's time the Plus plan got the attention it deserves.
298
178
59
4,163
416,284
Hasan Can retweeted
ANTHROPIC IS CRACKING DOWN ON VIDEO GAME REWRITES 🚨 Claude Code picked up a new rule around Oct 27, sent from Anthropic’s servers under a flag called "tengu_heron_brook". There’s no setting to turn it off. The full rule: “Do build original games, games in a genre, and games from the user’s own designs and assets. Don’t build a recreation of a specific existing game, even if it is renamed or called original.” Is this the end of game rewrites, modding and everything in between?
839
517
585
9,184
2,430,778
Sonnet 5.5 costs 2.8x GPT-6.1 Sol per task on the API. On $20 plans, a third
12
9
2
383
13,754
关于Gpt-6.1-terra 消耗过高的真相: 原来Gpt-6.1-terra 隐藏倍率高达 1.9x,实际订阅里只比 Gpt-6.1-sol 便宜30%。(via:linuxdo) 再加上 6.1 Terra基本上是6.1Sol一倍以上的速度,所以体感上 6.1 Terra消耗的额度就是 6.1 SOL 的 1.5 倍以上
8
6
2
64
10,437
OpenAI just alienated ChatGPT/Codex Plus users What a sad day
47
28
2
699
55,041
Hasan Can retweeted
Tibo 说 6.1 Terra 比 Sol 多两倍使用量。 看 GPT Plus,6.1 Sol 大概是 2.2 亿 tokens 一周 那 6.1 Terra 大概是 6.6 亿一周 可是 Claude Pro 的 Opus 5.5 能到 7-9 亿一周,速度大概 90 toke/s Sonnet 5.5 额度更大,甚至速度能上 200 tok/s 全方位的碾压。
65
17
6
948
115,753
Hasan Can retweeted
Did Codex just DECREASE usage limits for GPT-6.1 Terra??? Peoplr started testing Terra 6.1 since its release and noticed something strange. The 5hr usage allowance apparently dropped from $7.50 to $5.28. That's almost a 30% reduction. TF!?? Terra was supposed to be the more affordable, usage efficient option, but if the actual allowance is lower, are we really getting more usage? Anyone else testing Terra 6.1 noticed the same thing? 👀
18
5
2
63
9,689
Hasan Can retweeted
Introducing Kimi K3.1 Pro: our most capable 3T-class open-weight model. 🔹 1M-token context, 2.8T total / 104B active params (MoE), Multimodal. 🔹 Full-project coding: games in Unreal, Unity & Godot. 🔹 Excels at parametric 3D modeling in Blender & CAD. 🔹 More reliable at finance work: Accurate modeling with hallucination down by 65%. 🔹 Matching the strongest proprietary models across coding, agentic and knowledge benchmarks. Kimi K3.1 Pro and K3.1 Pro Highspeed (2–3× faster) are now live on Kimi.ai, Kimi Work, Kimi Code, and the Kimi API. Open weights by Nov 4, 2026. 🔗 API: platform.kimi.ai 🔗 Tech blog: kimi.ai/blog/kimi-k3-1-pro
1,301
3,243
2,384
30,318
4,297,352