Yasser

Technology x Society | Privacy, autonomy and trust. 🇵🇰 🇦🇪

[Yasser @earlgreyco]

Yes, there is. Treat people well and they'll stay.

[jack @jack]

there is no moat

[Yasser @earlgreyco]

What's actually sad to see: the internet reduced to a shady shitshow, or stupid sensationalism (thumbnails, and influencers) as signals to surface something new.

[Chubby♨️ @kimmonismus]

This is such a shame and so sad to see. @Kimi_Moonshot ruined their launch themselves. K3.1 pro is a great model and it did not need people to fake good reviews. But since they were trying to cheat the game it makes the whole launch look bad.

I did not receive this briefing and was not paid by kimi. But I can assure you, if I had known earlier that they tried to cheat the game and do illegal influencing, I wouldn’t have said one good thing about their model.

Super sad to see. Hope @Kimi_Moonshot steps forward, apologizes, and promises never to say anything like that again, because reputation is what matters most.

[Yasser @earlgreyco]

Nice little payday for everyone but the user.

[Watcher.Guru @WatcherGuru]

JUST IN: 🇺🇸 Judge orders $META to pay $25,000,000,000 for misleading users about their privacy.

[Yasser @earlgreyco]

@bznotes their burghul salad is top notch too.

[Yasser @earlgreyco]

Interesting experiment on the negative space between known parameters. unlikely to have resulted in the niantic game or flower or tetris.

[RE:Skob @re_skob]

Interesting idea. Had tokens to burn so I got Opus to put it to the test

[Yasser @earlgreyco]

At some point, when they stop trying to recreate the fucking Jetsons or Jarvis (take your pick), we will see a new class of software with useful intelligence baked in.

[Haider. @haider1]

i'm sorry but it's genuinely frustrating that these companies have some of the smartest engineers in the world building the smartest AI models ever

they could build almost anything, yet somehow we keep getting the same chatbot with a different model and wrapper

[Yasser @earlgreyco]

The $1200/yr GPT Pro plan has really gone to shit.
5 months ago, it was impossible to exhaust the weekly limit, plus the task would continue until complete even if it went to 0. Today, Astra is a no-go (ate 2 weeks worth in a day) and 6.1 Sol barely lasts 2 days.

[Yasser @earlgreyco]

@brucefenton Dynamic pricing is great in theory if capped at a MSRP. No one has implemented the consuner-friendly version of it yet.

[Yasser @earlgreyco]

Qwen 3.8 Max used to be top 5 on this, I think. They then shifted the goal posts with a new Index which pushed all western models up, chinese down.

[Artificial Analysis @ArtificialAnlys]

DeepSeek V4.1 Flash 1017 scores 45 on the Artificial Analysis Intelligence Index, improving on DeepSeek V4.1 Flash by 6 points, and is the equal #2 open-weights model

@deepseek_ai has released DeepSeek V4.1 Flash 1017, an update on their flagship model DeepSeek V4.1 Flash. On the first-party API, DeepSeek V4.1 Flash 1017 is priced at $0.30 per 1M input tokens and $1.20 per 1M output tokens during peak hours, with off-peak pricing half of peak rates, the same pricing as DeepSeek V4.1 Flash.

Key results:
➤ DeepSeek V4.1 Flash 1017 makes gains in agentic capabilities, knowledge and scientific reasoning. DeepSeek V4.1 Flash 1017’s Elo rating on GDPval-AA v2.1 is 1746, an increase of 146 points from DeepSeek V4.1 Flash’s rating of 1600. On AA-Briefcase v1.1, its Elo rating increases by 150 points from 1420 to 1570. On AA-Omniscience, its non-hallucination rate is 33%, a 29-point improvement from DeepSeek V4.1 Flash’s 4%, while seeing an improvement in accuracy from 43% to 46%, pushing its overall AA-Omniscience Index score from -5 to +5. On CritPt, DeepSeek V4.1 Flash 1017 scores 24%, up from 14% for DeepSeek V4.1 Flash.

➤ DeepSeek V4.1 Flash 1017 is now the #2 open-weights model alongside GLM-5.3. With a score of 45 on the Artificial Analysis Intelligence Index, DeepSeek V4.1 Flash 1017 is behind MiMo-V2.6-Pro by only 1 point and ahead of other open-source models like Step 5 Preview and Kimi K3 (both at 44).

➤ With a Cost per Task of $0.26, DeepSeek V4.1 Flash 1017 sits in the most attractive quadrant of Capability vs. Cost. With a Cost per Task of $0.26 and Intelligence Index score of 45, DeepSeek V4.1 Flash 1017 compares favorably to GLM-5.3-Flash at $0.28 per task and a score of 42. At the same time, MiMo-V2.6-Pro (46) at $0.13 per task and GPT-6.1 Sol (medium, 48) at $0.21 per task both offer higher capability at a lower Cost per Task.

Additional model details:
➤ Context window: 1M tokens (equivalent to DeepSeek V4.1 Flash)
➤ Pricing: $0.30 per 1M input tokens and $1.20 per 1M output tokens, unchanged from DeepSeek V4.1 Flash. Cache hit price of $0.006 per 1M tokens, a 98% discount
➤ Size: 552B total parameters (16B active)
➤ License: MIT
➤ Providers: Available through DeepSeek’s first-party API, @FireworksAI_HQ , @baseten, @DeepInfra , @wafer_ai , @Cloudflare and others

[Yasser @earlgreyco]

mentioned this just a few days ago.

[Matthew Prince 🌥 @eastdakota]

Amused that AWS just unilaterally cancelled Cloudflare's Reinvent booth. There are always signs when someone finally realizes they're losing. Guess we'll just attend and throw the most fun side events without paying Amazon! Reply if you want an invite.

[Yasser @earlgreyco]

Overvalued and over-funded companies over-hire, then look for reasons/scapegoats to downsize when their reasonable revenues don't cover their over-sized expenses or justify their valuations.

[kuldeep @ku1deep]

So many reactions to this so I have another one. In June this year Samel Chaudhary wrote an MIT sloan working paper that essentially said that in the last 10 years the net GAAP income generated by SaaS compnies was -$ 78 Billion after absorbing $350bn in paid in capital. Almost all of that went to Stock based compensation. SaaS companies mostly transferred value from public shareholders to employees through dilution.

Almost all of these companies are overstaffed. sometines by 3-4X. Bending spoons is just the invisible hand of the market arriving to repair the system. This will happen more and perhaps needs to happen. There are a lot of companies that are doing more than 100mn in ARR that can be run very successfully with <100 people and stay adaptable and deliver value to their customers.

[Yasser @earlgreyco]

@teovito I wouldn't use this perspective. You can get away with lower poly count figures if you use the TV perspective for your camera. Use these angles sparingly for replays etc if you ever do them.

[Yasser @earlgreyco]

Misclassification. You still need software; your chosen AI is taking over the cognitive load for you. All this means is you're paying in tokens, dollars and time to ensure you get the quality you want.

[Tim Ferriss @tferriss]

Kevin Rose (@kevinrose) on the coming “headless” era of software.

[Yasser @earlgreyco]

Hopeful post, hopeless comments section.

[Alexey Fateev @superalesha]

You distilled HUMANITY. You trained on our knowledge, scanned millions of books and threw the physical copies away.

You built a business on work people spent their lives creating, and charge us for access to the result. But when someone uses YOUR output to build a competitor, suddenly learning from someone else’s work becomes unacceptable.

That knowledge didn’t start with Anthropic. Training a model on it doesn’t make it fucking yours.

[Yasser @earlgreyco]

@Cloudflare is incredible; the @AWS of this era. More, in fact.

[Yasser @earlgreyco]

@teovito @deifosv animation should match gameplay mechanics:

- visual differentiation in shots.
- running animation
- power meters

Virtua tennis for the dreamcast is doable north star
https://t.co/TvyBoMYea0

[Yasser @earlgreyco]

"a thousand no's for every yes."

[Jun @NSJunXiu]

This video by Apple has aged like wine in today’s superintelligence era.

 2013

[Yasser @earlgreyco]

Preserve human-to-human. Enable AI-to-AI.
The incentive structure is what's moving this in the wrong direction.

[Luke Whiting @thelukewhiting]

asked @Muse to call @Walgreens to check stock, it went great

[Yasser @earlgreyco]

Single-handedly proving that good CX leads to runaway advocacy and deep loyalty.

[Tibo @thsottiaux]

Day 3/

The big one is GPT-6 in Chat, but today is also a little celebration day with a new high of 40M active users across Codex and ChatGPT Work.

Loading a banked reset in everyone's paid accounts. See you again tomorrow!

[Yasser @earlgreyco]

Is this convenience, or training people not to think?

[OpenAI @OpenAI]

GPT-6 and Intelligent UI, now rolling out in ChatGPT for everyone.

Intelligent UI in ChatGPT delivers fast, interactive answers that make everyday questions more visual, complex topics easier to grasp, and interactive tools for your task available on the spot.