|
I'm afraid most academic institutions shy away from teaching critical thinking, because if they did, it would enable the students to see through their B.S..
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
An important aspect to getting the code right is getting the spec right. If AI/LLMs become good enough to generate code for large systems, the problem simply shifts from writing correct code in a computer language to writing a correct spec in a natural language. In that case, programmers should be taught how to spot ambiguous or incomplete specs and fix them.
|
|
|
|
|
A trio of ethicists at the University of Oxford's Oxford Internet Institute has published a paper in the journal Royal Society Open Science questioning whether the makers of LLMs have legal obligations regarding the accuracy of the answers they give to user queries. People don't listen to ethicists, so why should AI?
|
|
|
|
|
Article wrote: whether the makers of LLMs have legal obligations regarding the accuracy of the answers they give to user queries.
Yes, please.
M.D.V. 
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
But they have no problem with LLM makers sucking up shitloads of copyrighted content without consent? Then they are not ethicists. They are pundits.
Quote: The researchers also suggest that LLMs used in high-risk areas such as health care should only be trained on truly useful data, such as academic journals
Ha ha ha ha! Ha ha ha ha ha! Ha ha! Ha ha ha!
|
|
|
|
|
I think the responsibility should lie with the person using the AI information. If they rely on it improperly to make an important move, then it is they who should pay the price, not the providers of the AI.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
At the current level of AI, when they can do no more than act as aids to decision-making, I agree with you. What happens when AI models take into account so many factors that no mere human can second-guess them?
Freedom is the freedom to say that two plus two make four. If that is granted, all else follows.
-- 6079 Smith W.
|
|
|
|
|
To respond to your question, I believe that judgement is still uniquely a human ability that I don't see AI having at any time in the future.
The difficult we do right away...
...the impossible takes slightly longer.
|
|
|
|
|
Yes and no. When I asked Bing Search for results of the New Hampshire primary it flat out lied to me.
|
|
|
|
|
Reddit executives discussed plans on Tuesday for making more money from the platform, including showing ads in more places and possibly putting some content behind a paywall. In the future, ads might have content attached
|
|
|
|
|
Kent Sharkey wrote: Reddit executives discussed plans on Tuesday for making more money from the platform, including showing ads in more places and possibly putting some content behind a paywall.
bb reddit 
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
I think I want them to just start charging. It would probably kill the site though.
Anyone want to make a reddit clone that has a mission of converting to paid access after reaching 1M daily users?
I vote we call it wispr. It will also be organically grown kinda like old school gmail rollout. Invite only.
|
|
|
|
|
Users are urging Microsoft to rethink how it shows sender email addresses in Outlook because phishing criminals are taking advantage, using helpful, friendly names to serve up emails loaded with malicious intent. At least that's what the email I just got from Bill Gates said
|
|
|
|
|
That's what mouse hover is for...
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Over the past six months more than 890,000 new packages (as opposed to updates for existing packages) were published on npm, of which between 613,000 and 667,000 – or around 70 percent – were Tea protocol spam, created in the hope of financial reward. Tea and spam do not go well together
But of course nothing actually goes with spam. Except those bloody vikings.
|
|
|
|
|
Better spam than malware  
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Spam actually goes pretty good with mac-n-cheese, if you add some good spicy salsa to it.
|
|
|
|
|
.NET Aspire is a new cloud-ready stack tailored for .NET, enabling developers to quickly and easily develop distributed applications. If you aspire to Aspire
Which no one outside of Redmond actually wants, but whatever.
|
|
|
|
|
Kent Sharkey wrote: If you aspire to Aspire
I do Aspire to be rich... but I do know I won't
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Researchers have demonstrated a method to bypass an anti-phishing measure in Microsoft 365 (formerly Office 365), elevating the risk of users opening malicious emails. It can't center a line of text vertically, but it can help hack your machine
Yes, I know "modern" CSS can center better, but my scars are still there.
|
|
|
|
|
Kent Sharkey wrote: It can't center a line of text vertically, but it can help hack your machine
Do you want help with that?  
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Kent Sharkey wrote: my scars are still there
An eternity for an entire generation! 
|
|
|
|
|
What I want to touch on in this post is the increased usage of auto and template type deduction that I’m seeing in newer codebases. Data types deemed handy: auto edition
|
|
|
|
|
auto var type = lazy and chaotic;
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
"Boss makes a dollar, I make a dime. Auto it is, all the time!"
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
