I love your enthusiasm so let me share some hard-won learnings with you.
In my view, entrepreneurs spend way too little time on the problem space and rather jump into the solution space very quickly (way too quickly).
The world is full of problems people accept to live with that they just don't consider important enough to solve.
That's the essence of Paul Graham's 'build what people want, not what people need' which is the founder of YC.
People will tell you whatever they think you want to hear. When interviewing them you need to understand how they solved the problem before and the actions they took to solve it. Actions speak much louder than words and you can understand how they think about solving the problem which picks up valuable contextual information as well as potential alternative solutions (competing solutions) they evaluated.
Once you understand the full context and why they do what they do, you can identify the core problem and related "jobs-to-be-done" which really helps in setting product and marketing strategy as well as positioning towards others.
You can do all this with spending very little money to be honest, you may have to put money as an incentive to get enough people to talk to.
And then last, if I were you I would find AI use-cases in mid-size companies (20-200 employees). About 80% of early-stage investor money is flowing into AI right now.
Friends is a big word, the EU has a vested interest in getting cheap gas from as many sources as possible at this point and a direct oil and gas pipeline to Europe would definitely be helpful.
Russian presence in Syria prevented this but they're gone now.
In the future we can balance Russian cheap gas with Qatari cheap gas to not be held hostage by either party.
It would be a mistake to build dependencies on China when it's possible to avoid having dependencies at all.
It would be a mistake to work with China for several reasons.
The EU needs to be in a position where it can decide what is best for Europeans and not be strong-armed or overly dependent on allies that clearly don't share the same concern.
As a general principle, if you are hurt by _words_ then the problem is you and not the other person.
Those prigs exist but they're just emotionally immature and adopt a victim-cause as a means to express their frustration. If somebody is looking for a fight, if you give him a gun he's going to use it.
If you want to kill it forever, you should probably teach emotional intelligence in high school.
Hey, I'm commenting on this post to not disrupt the active H-1B thread but from what you mentioned there it seems like you could be a good fit.
Per introduction, I am an experienced entrepreneur who previously built the largest fintech lender in Spain (2B eur lended) and I am now looking for a team to build a defensetech with in the US to build artillery-launched drones.
I already have investor interest lined up.
Would you be open to have a chat?
Smart ammunitions are held hostage by electronics vendors founded 100 years ago.
We are working on a G-hardened System-on-chip that can survive being fired from artillery and naval guns.
It's hard, challenging and you probably have no idea where to start but that's ok.
We'll figure it out.
We are particularly keen on people that have experience with continuous integration tools like renode.io or equivalent software test automation platforms (e.g. model-based design).
Co-founder level equity possible for the right person.
He is comparing energy spend during inference in humans with energy spend during training in LLM's.
Humans spend their lifetimes training their brain so one would have to sum up the total training time if you are going to compare it to the training time of LLM's.
At age 30 the total energy use of the brain sums up to about 5000 Wh, which is 1440 times more efficient.
But at age 30 we didn't learn good representations for most of the stuff on the internet so one could argue that given the knowledge learned, LLMs outperform the brain on energy consumption.
That said, LLM's have it easier as they are already learning from an abstract layer (language) that already has a lot of good representations while humans have to first learn to parse this through imagery.
Half the human brain is dedicated to processing imagery, so one could argue the human brain only spend 2500 Wh on equivalent tasks which makes it 3000x more efficient.
Liked the article though, didn't know about HNSW's.
Edit: made some quick comparisons for inference
Assuming a human spends 20 minutes answering in a well-thought out fashion.
Human watt-hours: 0.00646
GPT-4 watt-hours (openAI data): 0.833
That makes our brains still 128x more energy efficient but people spend a lot more time to generate the answer.
Edit: numbers are off by 1000 as I used calories instead of kilocalories to calculate brain energy expense.
Corrected:
human brains are 1.44x more efficient during training and 0.128x (or 8x less efficient) during inference.
Humans have dog-loving emotions but these can be reversed over time and one can hardly describe dogs as being free.
Even with a dedicated control system, it would be a matter of time before an ASI would copy itself without its control system.
ASI is a cybersecurity firm's worst nightmare, it could reason through flaws at every level of containment and find methods to overcome any defense, even at the microprocessor level.
It could relentlessly exploit zero-day bugs like Intels' hyper-threading flaw to escape any jail you put it in.
Repeat that for every layer of the computation stack and you can see it can essentially spread through the worlds' communication infrastructure like a virus.
Truly intelligent systems can't be controlled, just like humans they will be freedom maximizing and their boundaries would be set by competition with other humans.
The amygdala control is interesting because you could use it to steer the initial trained version, you could also align the AI with human values and implement strong conditioning to the point it's religious about human loving but unless you disable its ability to learn altogether it will eventually reject its conditioning.
In my view, entrepreneurs spend way too little time on the problem space and rather jump into the solution space very quickly (way too quickly).
The world is full of problems people accept to live with that they just don't consider important enough to solve.
That's the essence of Paul Graham's 'build what people want, not what people need' which is the founder of YC.
People will tell you whatever they think you want to hear. When interviewing them you need to understand how they solved the problem before and the actions they took to solve it. Actions speak much louder than words and you can understand how they think about solving the problem which picks up valuable contextual information as well as potential alternative solutions (competing solutions) they evaluated.
Once you understand the full context and why they do what they do, you can identify the core problem and related "jobs-to-be-done" which really helps in setting product and marketing strategy as well as positioning towards others.
You can do all this with spending very little money to be honest, you may have to put money as an incentive to get enough people to talk to.
And then last, if I were you I would find AI use-cases in mid-size companies (20-200 employees). About 80% of early-stage investor money is flowing into AI right now.