disambiguation

In reference to this post i saw a few weeks ago:

https://lemmy.zip/post/50193734

(Lemmy is a reddit style forum)

The author mainly demos their "custom tools" and doesn't elaborate further. But IMO is still an impressive showcase for an offline setup.

I think the big hint is "open webui" which supports native function calls.

Some more searching and i found this: https://pypi.org/project/llm-tools-kiwix/

It's possible the future is now.. assuming you have an M series with enough RAM. My sense is that you need ~1gb of RAM for every 1b paramters, so 32gb should in theory work here. I think macs also get a performance boost over other hardware due to unified memory.

Spit balling aside, I'm in the same boat, saving my money, waiting for the right time. If it isn't viable already its damn close.

Not my build and not coding, but I've seen some experimental builds (oss 20b on a 32gb mac mini) with Kiwix integration to make what is essentially a highly capable local private search engine.

Sorry that's not what i was trying to convey, but rather the elaborate loop hole to exploit cheap off shore labor over domestic workers. And yeah to your point about bad technical, plus my focus on the high powered hardware, all add up to legitimate safety concerns.

Surely the overlap between people with both the wealth and the preference for industrial machinery remote controlled by an underpaid worker from the Philippines in their house (around their children) over an organic house keeper is vanishingly thin, no?

The total addressable market for giant fighting robots on the other hand...

Semi off topic, but i always thought it would be cool if there was a kind of multiplayer for ebooks - threads and notes in the margins similar to how dark souls lets you write brief hints for other players that you can hide around the game world.

Its impractical due to the technical and legal challenges, but it would be neat to see the thoughts of other readers page by page.

I suppose intelligence can be partitioned as less than, equal to, or greater than human. Given the initial theory depends on natural evidence, one could argue there's no proof that "greater than human" intelligence is possible - depending on your meaning of AGI.

But then intelligence too is a dubious term. An average mind with infinite time and resources might have eventually discovered general relativity.

> people do have some ability to distinguish truth from lies

In a pre-AI era, this is a reasonable heuristic.

But I think everyone has their own internal barometer - how much trash are they willing to sift through for gold?

I'm concerned that AI will poison the well to such an extent that people stop trusting, and subsequently stop using, the web altogether. This isn't only a sad state of affairs for the average user, but ties into an existential risk for the business model of the entire tech industry.

Your point is valid, but i disagree with the framing.

The issue is the blind trust that the internet is built on top of.

It's was that initial good faith that made the internet a special thing - you could come online and discover all the weird interesting things people were up to, have conversations with real people that weren't possible in the real world.

At this point most platforms have figured out how to exploit and profit off the blind trust - but AI is threatening to annihilate it completely.

I'm not worried about the generic content, the worthwhile stuff is what's at risk.

There are no politically neutral institutions anymore, is the more general issue. "You're either with us, or against us." It becomes less about doing a good job and more about playing for the team.

Aging is related to shortening of telomeres - the speculated evolutionary advantage is that it's a mechanism to protect against cancer.

Unclear from the study what the stem cells are doing to address either problem.

An interesting take from EA a few days ago:

https://forum.effectivealtruism.org/posts/hGwAohPbwCrDT5ynY/...

There's obviously a lot to unpack, but IMO the buried lede - the new idea Thiel is trying to push - is that the "antichrist" wants to regulate AI.

This is of course a persuasion effort by Thiel that tries to do a few things - I don't have a personal opinion, but speculating I assume Thiel is trying to get ahead of something here. We haven't really seen a regulatory push in the mainstream _yet_. On the contrary, there has been a number of legal exemptions. I would bet (60/40) that in the near future we'll see "AI regulation" become a pertinent topic of mainstream discourse.

The effectiveness and pragmatism of economic theory aside -

It's interesting to observe that highly opinionated topics (such as economics) tend to follow a "team sports" pattern. More than day to day utility, people argue passionately for their "home team". It becomes a point of pride and identity.

Of further note is the role of education - generally considered a form of training to improve your personal wellbeing - yet it has a secondary function. It can convince you to "change teams."

rethinkeconomics.org says:

> The problem we face is the total dominance of one way of teaching, which promotes the marketisation of society, leading to increased inequality, injustice and significant harm to the natural world.

A cohesive society is one with "team support" parity between the people and government. It begs the question - what happens when society reverses parity?

> Others have looked more towards what a post-dollar world would look like: a gold-backed currency. A basket of commodities. A BRICS+ crypto-alternative. Bretton Woods 3.0.

> All of it’s noise — and none of it matters.

I've been reading libertarian de dollarization doom posts since 2008.

The question is, if the system is so messed up, why does everyone continue to put up with it? It's exactly due to the lack of viable alternatives. So long as the American monetary system is the only game in town, it will remain the reserve currency regardless of the debt or strength of the dollar.

Insufficient data for a meaningful answer. We don't really have a clear grasp of the relative benefits and harm yet.

Automobiles offer enormous gains in convenience and productivity, but at the cost of a non-negligible mortality rate as well as environmental impact. Society deems this a net-positive.

The two major unknowns with "AI companions" are: can people be trusted with this level of autonomy, and is the massive centralization of personal data going to result in abuse and exploitation.

I doubt any amount of discourse can answer questions of "how much", but perhaps the conversation can anticipate potential harm and "invent the traffic light" before the car crash equivalent becomes common place.

To do that, we have to first answer: what are the characteristics of the failure-mode of an AI companion?

Replying mainly to the title - but I'm surprised VR based 3D modeling never took off. I've only dabbled with blender, I know it's a powerful tool, but the learning curve just to navigate the interface is steep - compared to some creativity VR programs which felt instantly intuitive for 3D modeling. I guess for a professional digital artist, having fine technical control of your program is more important.

Settling for less can be economically optimal; planned obsolescence for producers (securing revenue), riding the bus for consumers (affordability).

Once it stops making obvious mistakes on a regular basis, then I'll take it seriously.

The paper linked in the article is paywalled, but after a bit of sleuthing I found a full access pre-print (I assume hosted legitimately)

https://osf.io/preprints/psyarxiv/d5fz2_v3

The section on measuring over-confidence is interesting, specifically how they address the shortcoming of self-reporting confidence (if you're bad at the task in the first place, you're also going to be bad at estimating your performance, where the act of estimating resembles the task itself).

Another fun activity is to visit the appendix (p. 47) which lists all of the True and False Conspiracies used in their assessment.

> We are all, collectively, building a giant, shared brain.

"Shared" as in shareholder?

"Shared" as in piracy?

"Shared" as in monthly subscription?

"Shared" as in sharing the wealth when you lose your job to AI?

Your math is incomplete. The whole formula includes interest rates and total amortized cost as well as the average / median income, cost of living, time to employment, and employment rates of new graduates. Additional exercises for the reader: find the average TTL of full loan repayment (~20 years) and the percent of graduates where debt exceeds earnings (~20%).

This isn't to say that degree attainment isn't statistically better on average, much like paying off ransomware. The price of tuition seems to reflect the lack of reasonable alternatives more than the intrinsic value of the education.

Feel free to unpack at your leisure, it's all in my original comment. I think it's both clear and concise, though you're welcome to be specific about what's left you feeling confused.

> US education and research is still one of our strong points, drawing in the brightest minds from the entire world.

The Average Joe graduates with enormous student debt and ends up in a low paying job, one that's entirely unrelated to their degree. They never experienced the benefits of education nor research. In fact, they feel quite ripped off by the whole affair, so this point holds no stock for them.

> So... Burn down academia with no replacement? Shutter and board research and leave America with no future?

This is a bit of a hysterical reaction to "budget cuts." Academia isn't going anywhere and probably won't change much, it's just going to have less money. In fact, k-12 education is seeing far more severe budget cuts as part of the bill, but there are fewer prophecies of doom in its honor. Academics argue passionately for their own bread, but not for those who are starving. (They also send their kids to private school)

The institutions are elite. So are the people who study there, and so are the decision makers directing the goals and strategies behind the research.

This feels like a spectator sport to me - elite classes fighting amongst themselves while ordinary people are caught in the crossfire.

Sure, I do feel bad for the well meaning professors, scientists, and researchers (some of whom are personal friends) that are losing their grant funding - techies of all people should be sympathetic to a sudden disruption in funding. But if the elites are asking us to take sides, they need to make a stronger case that they're really on _our side_ - that requires more than appealing to mutual self-interest. Academia must reconcile a litany of shortcomings and broken promises.

For the promise of education, we see a large and growing contingent of debt holders outweighing the number of available employment opportunities. Tuition is at an ATH, yet more and more students are walking away with worse outcomes. A grim forecast for the younger generation.

For the promise of research, we've seen a shift from truth-seeking to grant-seeking, a punitive culture of ideological conformity, and a pipeline to brain-drain the rest of the world, favoring foreign talent over local. Hardly the unbiased bastion of free-thought we've been taught is the foundation of good science.

As for ideology, I'm not sure the intelligentsia has ever been neutral, if that's even possible, or if neutrality itself isn't another ideology. As far as I can tell, this is just the pendulum swinging back. Academia is experiencing the consequences of its failure to live up to its promises to society and hold itself accountable for the negative externalities of its own agenda.

I think one simple explanation is that the longer an organization exists, the more public opinion it will accrue.

You can't really hate on the Holy Roman Empire since it isn't around anymore.

For the curious:

https://en.wikipedia.org/wiki/Censorship_by_Google

https://en.wikipedia.org/wiki/SafeSearch

https://en.wikipedia.org/wiki/Search_engine_manipulation_eff...

There is a definite psychological component to depression - CBT and meditation are effective treatments - yet a point often absent is how one becomes depressed in the first place. We tend to moralize a lack of resilience and a bad attitude. A more empirical perspective is to look at studies using animal models which show that depressive symptoms can be induced via factors like chronic stress, social defeat, and general deprivation of wants and needs.

https://pmc.ncbi.nlm.nih.gov/articles/PMC8111503/

I'm concerned by the pervasive thought pattern held by many, but especially engineers: that you can think your way out of any problem. This bias leads to preferential seeking and over-prescribing of therapeutic treatments.

Yet how many people who self report depression are also satisfied with the various external factors in their life? Perhaps the correct framework is: a healthy environment is the cornerstone of a healthy mind.

I'm sure it's possible with enough meditation and medication to achieve enlightenment whether you're exiled in the Siberian wilderness or locked up in a castle dungeon, but to do so is working against the grain of nature. Unhealthy environments have a negative impact on individual health. The Foxconn "safety nets" come to mind.

Though I could be mistaken, perhaps zoo animals are only depressed because we haven't fully translated "mindfulness" into ASL.

"smart" could be defined as: the act of consistently making good decisions - which can further be defined as: effective optimization towards an outcome. (defining the outcome is itself a matter of making a "good decision")

This requires all of: being aware of a given problem, being sufficiently informed of the relevant context (which is further a matter of curiosity, discerning between trustworthy sources, and robust sense making), and finally caring enough to apply any attention and effort to the issue in the first place.

In this regard, almost everyone is "stupid' about everything most of the time. If anyone manages to achieve "smartness" it's usually in a very narrow decision space.

In terms of AI and education, the problem is: the path of least resistance is an optimal one - at least in a greedy sense.

The usefulness of the tool and "smartness" of the user are irrelevant to the core issue - general education is rapidly eroding. This is strongly correlated to (if not outright caused by) the ongoing rapid changes in technology.

The issue is that structured education originally meant: relying on your own wits, which in turn strengthened them. No cheat codes allowed.

This is no longer the case. Not only because of students using AI, but because "the path of least resistance" applies to educators and administrators as well.

Technology will change but educating people remains a fundamental good - to that end, institutions must adapt to make sure every student gets the proper enrichment they deserve. Get cheat codes out of education.

Previous Attempt: https://en.wikipedia.org/wiki/Shoreham_Nuclear_Power_Plant

tl;dr it was commissioned, constructed, and closed due to local safety concerns in the wake of the Three Mile Island accident.

Interesting excerpt from the wiki:

In 2004, the Long Island Power Authority erected two 100-foot, 50 kW wind turbines at the Shoreham Energy Center site,[18] as part of a renewable-energy program.[19][20] At a ceremony, chairman Kessel stated, "We stand in the shadow of a modern-day Stonehenge, a multibillion-dollar monument to a failed energy policy, to formally commission the operation of a renewable energy technology that will harness the power of the wind for the benefit of Long Island's environment." The turbines generate 200 MWh per year, or 1/35,000th of the energy the nuclear plant would have produced.[21]

> Iran is a very minor threat to europe and US, and very minor threat in general.

Independently, yes. An axis is what's problematic.