@JackEllis $15 per month sounds very cheap compared to the time required to do all of these things 😊
Specialization has worked well historically for a reason, I guess.
@dhh It is still a tool though. And a tool does not, unlike coworkers, have responsibility for the result. Only humans can be accountable, which requires skill, judgment and experience - even if not a single line of code would be written by humans.
@forgebitz Crazy though, I know, but: what if the quality of software depends on the skill of the developers who make the software, and not just on the tools that they use.
@svpino Partly indicates that productivity is complicated. If some tasks are made more efficient, bottlenecks can appear elsewhere in the process which can put a cap on the actual productivity unlock (or nullify the efficiency gain if the bottleneck requires new/more work).
@forgebitz I think it’s broader than tech. Most people (including myself) are not entrepreneurs. There are probably far more book dreams than there are actual books. The audience is typically larger than the number of people on stage. I don’t think tech changes this by much.
@CaptainInsightX Being a human and all that it makes possible, including making decisions for which I am accountable, understanding systems outside of a context window, being responsible for systems, tacit knowledge, human collaboration. The list is long.
@DanielSmidstrup Well, someone needs to write the remaining 10% and also understand the 90% that was generated. And take ownership of 100% of it. AI code generation does not mean that the developer role has been automated.
@MuthaNagavamsi No one really ever had a clue about what’s coming in one year. Nobody knows for sure. People can make plans and projections, but everything is always based on assumptions and guesses. This includes ”<insert new tech here> changes everything” type predictions.
@pmarca No problem, because the AI models serve us humans and we ultimately decide which long term plans and ideas that are used. Humans are also the arbiters of taste.
Since AI models are tools rather than competitors, humans benefit as the models become more capable.
Magical OpenClaw experiences that use frontier models cost $300-1,000/day today, heading to $10,000/day and more. The future shape of the entire technology industry will be how to drive that to $20/month.
@Afinetheorem The truth is that nobody knows exactly how the job market of the future will look like. Everyone is just guessing. So any detailed prediction that is expressed with certainty should be seen for what it is: a guess among many.
Tech changes nothing by itself.
People create change, not the tools that we use.
It is not demos, benchmarks, tech specs or technical capacity that determine humanity’s future. It is, and has always been, determined by human choices.
When are tech folk going to get that people like wasting time, it's life. They don't optimize for efficiency, they try to get by, they watch dumb stuff, they enjoy shopping. Inefficiency is another work for living and life.
Your m mean and median job isn't a software engineer in Menlo Park, it's Ashley in accounts in a not for profit in Columbus, it's Jesse , the office manager for a tool rental business in Tallahassee, they are more likely to use a Fax machine than Slack.
They quite like meetings because they like chatting, they'll use AI to make a better invite to their baby shower, not agentify their job.
These people, nor their bosses boss, aren't in a rush to build software as a side hustle, they are keen to use AI to check if their vet is overcharging them. They'd like AI to check spelling on the email to the school governor.
They don't want agentic commerce, they want AI to be in the background and make living a little less stressful
@tomfgoodwin Many predictions are too heavily based on technical capability.
To make credible guesses, it’s crucial to understand _both_ tech _and_ people.
It is probably even more important to understand people than to understand the tech.
@SecretCFO I think an underestimated benefit is that spreadsheets also provide a well-known way of doing things.
It comes down to how people get work done each day. Predictability and transferrability matter. Efficient, predictable routines have staying power.
@andrewchen I think a big reason why spreadsheets have staying power is that they provide a predictable, transferrable and well-known standard for getting work done.
Custom AI-generated software has different kinds of benefits. So spreadsheets will likely endure.
@Noahpinion Why, though? LLMs are computer programs - tools - that humans control. We are no more controlled by LLMs than we are controlled by cars or power tools. The only scenario in which I see this changing is if we get to a Skynet type situation, but that is science fiction.
@corbin_braun Even if anyone could, doesn’t mean that they will. To use an analogy, it is likely not the lack of pens, papers and word processors that prevents people with writer ambitions to write the next great novel.