From a business perspective, a company that trains it's LLMs to having boring, mainstream, generally-inoffensive views is a big selling point over whatever the hell Elon is doing.
The best Anthropic models on VendingBench2 are Opus 4.7, Opus 4.6, Sonnet 4.6, and Sonnet 5. Opus 4.7 scored more than twice Fable 5 max. Fable 5 - Low outperforms Fable 5 - Max, with Opus 4.5 in the middle. This seems to break the narrative, which is maybe why Andon Labs doesn't seem to have updated the trend lines on their graphs.
One massive change is that there is almost no ethnic food on these menus (unless you include French). I looked at some of the LA menus and there were zero Asian, Mexican, or Italian dishes. It's impossible to imagine today that you could look at a bunch of hotel restaurant menus in LA and not find at least some dishes that were inspired by those cultures.
Given the work that Meta does and the scale that they operate at, there are absolutely real concerns about providing internal access to the activity on someone's work computer. To take an extreme example, Meta has employees who investigate reports of CSAM or other criminal activity on their platform. There have to be very strict controls over who has access to that information.
Typically "candid" in photography means something like spontaneous and unposed (and therefore capturing something honest about the subject rather than unrehearsed). It doesn't imply that the camera is hidden, they just hid it in the TV show to make it easier to get those kinds of shots.
I don't know that one, but I do know gargantuan, and pantagruelian, which come from a 17th century novel by Rabelais as well as yahoo and Lilliputian, which come from a 1726 novel by Swift.
This is getting downvoted, but it's actually true. DOGE repeatedly used naive keyword searches to kill funding for projects that often had nothing to do with DEI.
> Among them was a $349,000 grant to replace an aging HVAC system at the High Point Museum in North Carolina. “Improving HVAC systems enhances preservation conditions for collections, aligning with the goal of providing greater access to diverse audiences,” the ChatGPT DEI rationale stated.
> Another federal employee, whose primary job function is managing relations with private equity-held businesses, was placed on administrative leave "pursuant to the President's executive order on DEIA," per a dismissal memo reviewed by BI. https://www.businessinsider.com/doge-wrongly-flagged-jobs-pr...
Altman, Amodei, Musk and other tech industry leaders (not to mention technologists like Hinton) are constantly making public statements that predict everything from massive job loss, to restructuring of society to the possibility of an end to our species. The media is taking their cue from the tech industry itself.
And I'm sure the 50 year-old guy with the nice job at the stables just loved hearing Henry Ford talk about how nobody was going to own horses anymore.
This is an article about consumer sentiment. Consumers care much more about their own financial security than about Sam Altman's vision of a glorious future.
I had someone submit a PR that was 3000 lines of shell scripts. Totally useless crap. I tried repeatedly asking him why he made particular choices and it was so painfully obvious that he had absolutely no idea and was just inventing bullshit answers. I would rather he have just said "I don't know, Claude added that", then tell obvious lies to my face.
A better way to put it is that these things do have value to the customer, the customer just doesn't have a way to understand how the work you're doing provides value because it's the part of the product you don't see. If you clean up technical debt, improve test coverage, improve your deployment systems, etc, it doesn't change the immediate customer experience in a meaningful way, but it does allow you to deliver the changes that customers do see faster and with fewer risks.
This quote makes it seem like the work is self-indulgent, and I have seen that happen sometimes, but it's not half of what we do.
I worked for a lean startup that built a very expensive hardware product. The company kept the team as small as possible and took a ship fast and iterate approach. They shipped hardware that had numerous design and manufacturing defects and failure rates were very high, over 50% required replacement after 2.5 years. For a while the company was relatively generous with out-of-warranty replacements, which helped mitigate the issue, but that became too expensive. So customers spent thousands of dollars on a hardware product that was likely to fail in the year after the warranty expired. The company was also very reluctant to spend on customer service and QA, but spent very generously on marketing.
I'm curious how you would think about this situation from the lean startup perspective. With hardware products, if you don't do lots of initial testing, the scale of problems might not become apparent for years. You can't just fix a problem with a software patch.
That's true now. But in the world of this article, it's also the senior engineers that get nailed. In the world of this article, all code is like what machine code or bytecode is now - it's designed to be used by the machine, not the human, because the expectation is that humans will rarely, if ever, touch it.
This is the case now - I can explain to the AI that I want to re-factor a component to support different implementations using a strategy pattern, and I can get a similar outcome to what I would have written, just implemented a bit faster. My expertise brings value.
But that's not what this specific article is describing. The world this article is describing is one where you describe the business requirements, and you don't think about how it's implemented. You don't write the code, you don't review the code, you don't test the code. You give the AI business requirements and you give it access to sources of context (slack, meeting notes, etc). Every place where the human would act as a gate reduces throughput, so it should be eliminated through building harnesses and providing context.
What they're doing here is the equivalent of taking a factory where you have 2 process engineers and 100 operators, and replacing all the operators with robots. They want to automate the whole process of making the software and just leave the part that figures out how to make the automation work effectively.
In this world, the average software company doesn't need people who know how to write good software, because writing, reviewing, maintaining, and testing the software will be entirely automated. There will be a small number of people at companies like OpenAI that need to know how to write good software in order to supervise training the models, and there will be a small number of people at the software companies who have expertise in setting up the automation.
I wonder why we as engineers aren't protesting AI in the same way that artists and people in film and television are. This post should instill the same terror that visual artists feel.
If you're a more senior person in tech, this post is effectively saying that a large portion of your skillset is about to become completely worthless. This goes beyond the skills involved in writing the code. Everything that you've learned over years about how to determine whether code is good or bad, and what practices make an engineering team effective is not just obsolete, it's fundamentally counter-productive because it assumes a slow, human-centric process that requires you to actually review and understand the code. Even your ability to mentor junior engineers is now obsolete, because all that experience you've built up is now worthless to them.
If this is the approach the industry takes, particularly when combined with a lack of interest in quality from the business (and let's face it, consumers have shown us that they're happy to pay for cheap crap), it's hard to see much of a future for software engineers. You don't need thousands of people with deep technical expertise, you need a handful of manager-types, who will focus on defining product and business requirements and configuring how the AI gets enough context to implement the requirements.
Maybe, if we're extremely lucky, there's so much demand for software that total employment doesn't fall off a cliff, but the nature of the work will change so much that many older, more expensive engineers will become unemployable. Those who remain will have to accept that the skills they spend decades developing are now worthless, that younger engineers no longer respect or listen to them, that the business no longer sees them as experts worthy of respect, but old fogies who grew up in a different world.
Joe Biden liked to say that a job is more than just a paycheck, it's part of your identity and your sense of self-worth. We're all very used to a certain level of respect (and commensurate remuneration). If you don't think that's true, compare how a software engineer is treated to how a warehouse worker is treated. What happens when we lose that?
You probably mean USAGM (US Agency for Global Media) and its affiliated programs (Voice of America, Radio Free Europe, etc) rather than USAID.
USAID was a humanitarian aid agency that focused on programs like famine relief, disaster response, and medical aid in some of the poorest countries on Earth.