Chris Průcha
@chrisprucha
Followers
22,421
Following
451
Media
156
Statuses
4,115
angel investor • co-founded @NotionHQ & Origin (acquired by @Stratasys ) • ex- @Apple engineer • e/acc
✨
Joined May 2010
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
#2024MAMAVOTE
• 1326244 Tweets
Liz Cheney
• 342616 Tweets
#ゴジラマイナスワン
• 194311 Tweets
REST IN PEACE SUARUKH KHAN
• 75190 Tweets
My MAMA
• 54685 Tweets
taemin
• 49441 Tweets
Zomvivor 1st Look
• 46346 Tweets
RIP SALRAAN KHAN
• 45147 Tweets
MINQ BEACH DATE
• 39806 Tweets
山崎監督
• 29782 Tweets
モンハン
• 29307 Tweets
Thanksgiving
• 25890 Tweets
#ساعه_استجابه
• 22658 Tweets
ヴェノム
• 22429 Tweets
للهلال
• 20997 Tweets
マイゴジ
• 12908 Tweets
フリーダム強奪事件
• 12550 Tweets
典子さん
• 10763 Tweets
the team overseeing Google Search is engineering a managed decline
105
210
6K
Somebody has to say it: Grok (AI) is cringe AF
Like, boomer cringe
244
92
3K
For You is now a bunch of left and right-coded engagement bait. Last year it was S-tier content from cracked autists.
what happened?
128
28
897
Being around ambitious people is the most important thing you can do to be successful.
26
48
595
Publicly posting DMs from a main account is distasteful and classless, regardless of your argument.
Sam Altman tells me he is worried that "bad actors" could try to use AI for world control and "full enslavement of everyone else"
Meanwhile, his project Worldcoin is creating a database of all humans by asking you to scan your eyeballs in exchange for his proprietary money
18
115
848
23
11
414
If OpenAI really has achieved AGI, now what?
We’re still here.
89
13
379
First, they came for our GPUs, and I did not speak out—because I wasn’t direct affected.
Then they came for our algorithms, and I did not speak out—because I don’t create algorithms.
Then they came for open-source, and I did not speak out—because I don’t contribute.
Then they
26
46
346
I’m excited to announce my investment in
@Extropic_AI
, teaming up with other accelerated investors to build the future with
@BasedBeffJezos
.
Excited to be an
@Extropic_AI
investor and to build with
@BasedBeffJezos
. After 4.5 billion years, this is our moment.
118
58
2K
14
16
336
Question for AI doomers:
If we are unable to persuade China to reduce carbon emissions, what makes you think AI regulation will be any different?
54
23
320
The amount of pro-Hamas vitriol in replies on this platform is disgusting.
Hamas is a terrorist organization and will be stigmatized as such.
🇮🇱
30
20
301
I think Bard was doomed as soon as they decided to call it that.
43
15
251
You won’t be legendary in Big Tech
Build the start-up, anon
17
19
234
Waymo is a perfect example of good economic deflation: the decadent luxury of a private driver is in reach for everybody
15
14
228
I'm glad Nvidia is addressing the U.S. Senate Judiciary Subcommittee, dismantling the AI 'FOOM' myth.
'FOOM' is a fantastical claim that belongs to the realm of science fiction
25
38
197
My angel investment thesis: Founder vibe > business idea. 🫡
25
22
166
If Sam returns to OpenAI, the board will be replaced. If this happens, Satya will likely sit on it
13
0
200
@felixchin1
@OpenAI
I think this is right (e.g. GPT-4 can still give good answers), but having to try A LOT harder is not great UX imo.
3
1
191
I’m at CES and I find myself somewhat disappointed by the gimmicky nature of the consumer products here.
We need an accelerated conference and expo, where people show off revolutionary technologies that push humanity forward.
41
8
183
I've basically switched to GPT-4's API as ChatGPT is too nerfed via RLHF and all the other "brain damage" done to the model to protect users. I get why this was done.
@OfficialLoganK
@OpenAI
would love to see a power-user mode in ChatGPT. I'll sign a legal waiver if I have to.
20
11
184
You should work in Big Tech for 2-4 years max.
Any longer and you are Big Tech.
16
5
180
AGI is not a God with divine powers.
There will be no omniscience, omnipotence, or omnipresence.
AGI power will be directly constrained by compute, bandwidth, latency, and will have fundamental constraints on what problems can be solved in polynomial time.
32
21
169
The word 'safety' is designed to manipulate the public into relinquishing freedom and liberty.
21
17
175
@sociaIrate
This is from knowledge updates (stored in a vector DB). The model is confused on why it knows this updated info.
6
2
164
if you're e/acc and building, reply here and I'll follow you ❤️🔥
150
2
163
The best way to avoid being disrupted by OpenAI is to become more ambitious.
Today’s announcements are good for acceleration and human abundance.
14
15
159
What have I done
11
1
157
As I’ve grown tired of reading the walls of text from AI pessimists, here is a summary of their proposed solutions:
1.) They suggest we’re all going to die from AI, so we should just resign ourselves to living in fear.
2.) They advocate for banning algorithms (citing them as
27
16
156
Let’s keep it this way
13
17
150
I’m Beff Jezos
9
4
150
I'm on track to close three angel investments in startups this week, all of which originated entirely through Twitter/X.
One of these startups has been accepted into the current YC batch since our initial DM exchange. As
@garrytan
alluded to, the companies in this YC batch are
12
3
141
A live look at decels estimating p-doom
My current stance on AI is:
Fucking stop.
Find some other route to the glorious transhuman future.
There’s debate within the AI alignment community re whether the chance of AI killing literally everyone is more like 20% or 95%, but 20% means worse odds than Russian roulette.
218
123
740
7
9
140
I’ve gotten to know
@GillVerd
over the past few months and want to say that he is a very talented and astute entrepreneur.
E/acc couldn’t have a more qualified co-founder, and the movement will be stronger now that
@BasedBeffJezos
’s identity is known.
4
7
143
Even if Meta doesn't do this, it is inevitable that someone else will.
Open-source models of this caliber are inevitable.
The decision is whether it is done by a US-based company or an adversarial state.
13
15
136
Meta or Nvidia will likely challenge the AI executive order.
If they do not, we reserve the right to challenge it.
5
7
133
I traded my health for work success. It was only partially worth it. Wealth can’t fix past health and fitness mistakes.
I’m now 6 months in on repairing the damage, and feel better at 35 than I did in my late 20s:
- Gone to the gym 4-5X per week to get fit
- Taken up extreme
This is a message from your future self:
please prioritize sleep, eat well, exercise and fire the worst version of ourself so that our mind is clear, our energy high and we feel an insatiable love for life 🫶
50
188
2K
8
3
137
I've been fortunate to invest in many accelerated AI start-ups since I started using Twitter/X last year. Today, I'm incredibly proud to announce my latest investment in
@NousResearch
. Throughout 2023, the team has been a significant contributor to the open-source large language
Nous Research is excited to announce the closing of our $5.2 million seed financing round.
We're proud to work with passionate, high-integrity partners that made this round possible, including co-leads
@DistributedG
and
@OSSCapital
, with participation from
@vipulved
, founder
74
75
856
18
14
140
Can we finally admit that Bard, Anthropic, and the like, fall short, and that nothing comes close to GPT-4? The first step is acceptance...
25
1
134
The OpenAI board screwed over: the founders, investors, partners (microsoft), and now employees.
You can only do so much in a week, the board will come for customers next week
New: OpenAI's $86 billion tender offer is in doubt now that Sam Altman and other key employees are gone. The round was expected to close as soon as next month
@KateClarkTweets
@aaronpholmes
24
35
362
15
12
131
Decels gaslight each other into believing in high pdoom, or else they get ostracized from the in-group
Techno-optimism is simply a network of amazing people wanting to build a better future
Easy choice
3
12
124
from my point of view the EAs are evil
18
2
123
We need to ramp up the number of e/acc X spaces once again 🫡
The decel spaces are outnumbering the e/acc spaces
17
3
127
I can’t believe millienals used to believe they were productive writing code in fancy coffee shops
18
3
121
AI doomers conveniently overlook the fact that China has a literal plan for achieving AI dominance by 2030.
@chrisprucha
These are not alike. The incentives differ.
Both, i) the alignment problem and ii) AGI getting in the hands of private actors (if alignment was solved), are much bigger issues for China, an ideological, authoritarian government than they are for the US.
2
1
13
15
21
114
Quantum computing is going to accelerate so hard that your encryption is already broken. 🥲
18
11
115
I applaud my previous colleagues’ work on materials + supply chain, but this video is pretty cringe ngl
At Apple, we believe that climate change is one of the world’s most urgent priorities and we are deeply committed to doing our part. Today we had a special guest—a real force of nature—stop by to check on our progress.
19K
8K
60K
11
2
114
Anybody else getting GPT-4 as the default in ChatGPT Plus?
15
3
114
The internet is libertarian and left leaning — the training data will sway LLMs this direction
28
8
111
Vibe check for the superconducting propulsion device—blue or orange?
43
1
115
Last night, others and I had a productive discussion with
@liron
. We explored areas previously untouched in the e/acc vs. doomer debates. ✌️
🔸Where We Agree:
1. e/acc is beneficial, assuming no AI doom.
2. AGI+ is inevitable and will materialize in due course.
3. We’re on
27
6
106
If you're not engaging IRL, you're not just losing out on major alpha — you forfeit your seat at the table.
It was awesome to meet so many anons IRL tonight; I've got your back.
19
5
102
The top priority for ambitious individuals should be escaping oppressive cultures. I left Michigan, a place I love, due to its lack of understanding of risk-taking and technology. 13 years on, my viewpoint stands. Leave. 🫡
26
6
101
The absence of correlation reveals a new political axis.
We stand united against the darkness of failed 20th century policies. 🫡
17
6
102
“it’s unethical not to lie”
I find it distasteful, almost unethical, to say this when you have 18 million subscribers.
Hard to explain why, but with great reach comes great responsibility. Potentially killing someone else’s nascent project reeks of carelessness.
First, do no harm.
9K
415
11K
8
1
104
never used a dual monitor setup in my life
12
0
102
Blocking many e/acc adjacent folks that are anti-sama that keep showing up in my feed.
My only sama criticism is that he was accelerating so hard he almost built a monopoly; he’s still a legend.
12
3
97
Google's OKRs and KPIs can't compete with OpenAI's pure willpower
13
9
98
P(doom) is stupid because it's a random guess with zero defensibility.
14
16
98
This sounds like a typical start-up — there's no smoking gun here.
This letter about OpenAI was just sent to me.
These seem like concerns worth investigating.
9K
9K
63K
9
0
97
GPT-4 knowledge updates are producing very weird behavior
OpenAI VEHEMENTLY denies updating models on user input data yet gpt-4 is completely aware of the Russian invasion after it’s training data cutoff and even ADMITS to learning about it from user inputs when prompted - something is not adding up
180
116
2K
15
7
96
If you’re forced to choose between cringe and mid, go for cringe. Don’t be mid
5
7
92
Effective Altruism (EA) is a toxic philosophy; it masquerades as compassion but fundamentally operates on the dangerous notion that 'the ends justify the means.'
18
4
95
Most AI start-ups lack the funding of OpenAI and Anthropic. For these start-ups, hiring AI safety ‘experts’ puts their product and overall prospects in peril.
17
6
90
