Jump to content

Talk:Lex Fridman

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Is Wikipedia biased...?

[edit]

I'm not a huge Wikipedia user, but I find myself reading Wikipedia articles often enough, usually after googling a certain figure or event. But I have started to notice Wikipedia articles and stated 'facts' seem to be leaning in a certain direction...


It seems to me that Wikipedia, something trying to emulate an encyclopedia, would try it's utmost to remain as impartial as possible - would ideally refrain, as much as possible, from including editorials or personal opinions; or if unable to refrain, would make sure to include enough of all sides of the issue to maintain its impartiality. This would be the BARE MINIMUM needed to be considered a valid source of information (like an encyclopedia) and not an Editorial. But in this, as well as many other recent articles, there is definitely a left-leaning bias (IMHO).

A simple example - in the list of Lex's interviewees, some more prominent individuals were left out - renowned podcaster Joe Rogan, presidential adviser Jared Kushner, current presidential candidate Robert F. Kennedy Jr, to name a few. Yet for some reason Mohamed El-Kurd is included in the list containing authors,professors, leaders of industry, actors, famous entrepreneurs, Prime Ministers and intellectuals.

Why?

Why can't something as ideal as an impartial encyclopedia and source of genuine information - remain just that? 2A00:A041:3CDC:8C00:DC14:DA63:E83E:13BF (talk) 21:13, 14 February 2024 (UTC)[reply]

We cannot list everyone who has been on his podcast. Please see this archived discussion; the consensus is to only include guests who have been written about in secondary sources. Primefac (talk) 21:16, 14 February 2024 (UTC)[reply]
it is hard to avoid noticing a certain lack of balance
Wikipedia at least is very clear and obvious in the censorship it executes al across the board. No room for wees "certain lack of balances" Lavidavi00 (talk) 08:27, 29 July 2024 (UTC)[reply]

Reception Section

[edit]

The following sentence should be removed from the reception section. I'll do it myself but I'm giving the few users determined to boost a particular perspective time to argue otherwise. The rest of the section is fair, though it should probably be reordered and expanded.

According to research experts interviewed by Business Insider in April 2023, Fridman "lacks the publications, citations and conference appearances required to be taken seriously in the hypercompetitive world of academia".

This statement has numerous problems. First, it is vague and subjective. What does "taken seriously" mean? This is just an insult. Fridman is currently employed by a university as a research scientist. Does this qualify as being "taken seriously" by academia? Is being a tenured professor the only way to succeed in academia? There's no way to make this determination because the statement is just an expression of subjective derision in a WP:BLP.

It asserts that he lacks "publications, citations and conference appearances". According to Google Scholar, Fridman has ~2000 citations. This is not exceptional but it's not nothing, and given that context the statement misleadingly implies he barely any.

The attribution of this quote is questionable. First, it is anonymous. Secondly, what is a "research expert". This is not a term used in academia. Also, the article attributes it to "experts" plural, but it reads like a quote from a single person. WP:BLP can do better than a questionable anonymous quote in an attack article.

The language of this sentence violates WP:BLP in multiple ways. It lacks precision and uses loaded language, when the policy requires that a person be commonly described that way in reliable sources. I am not aware of any other source that describes Fridman this way, other than the Business Insider article, which seems solely devoted to make him look as bad as possible. Chase Kanipe (talk) 11:29, 19 March 2024 (UTC)[reply]

Chase Kanipe – this wasn't an "anonymous quote" (your words in your edit summary). It is a quote that can be attributed to Julia Black. It isn't up to you to evaluate which critique should not be included just because you do not like it, nor for you to make up your own analysis or arguments about the number of citations he has had. Likely compared to peers his own age, his research output is not impressive. "solely devoted to make him look as bad as possible"?? Hmm that's your opinion (!) and it is starting to look like you are WP:WHITEWASHING. Wikipedia allows for inclusion on sources deemed reliable, and this source has already been discussed on a noticeboard where consensus was gained that it was acceptable. Zenomonoz (talk) 01:17, 22 March 2024 (UTC)[reply]
You seem to think that a source being deemed reliable warrants the inclusion of any statement from it. It is not the job of WP:BLP to aggregate every negative statement made about a person in a reliable source. Doing so can result in an article that violates WP:BLP or WP:NPOV, as I have argued is the case here.
You misunderstand my objection. I did not argue against the inclusion of the Business Insider article, as you alleged. As I said before, the other uses of this source in the article are fair. I argued against the inclusion of this particular sentence, because it is an imprecise expression of derision that probably violates both WP:NPOV and WP:BLP.
You have described this quote as "criticism". If it was a criticism, I would not object to including it. It's not a criticism, it's an insult. I ask you again, what does "taken seriously" mean. This phrase is more than an assessment of his research record, it derides him as a person. If this derision was repeated in multiple reliable sources perhaps it would be notable enough to warrant inclusion, but the judgement is only expressed in a single source and is sourced to anonymous "research experts". As written it fails to conform to an impartial encyclopedic tone.
You have stated that "it isn't up to you to evaluate which critique should not be included". Wikipedia is not an indiscriminate compilation of information. You're correct it isn't up to me, it is up to a consensus of Wikipedia editors to make considered judgements about how to conform articles to Wikipedia policy, which is why I started this conversation here and given reasons this sentence is not up to standard. Chase Kanipe (talk) 12:57, 22 March 2024 (UTC)[reply]
I argued against the inclusion of this particular sentence – that isn't a convincing argument. It's just you doing your own analysis, which isn't how Wikipedia works. If it's a quote from a reliable source, it's allowed to be included. Zenomonoz (talk) 23:22, 22 March 2024 (UTC)[reply]
That is incorrect. Wiki's rules are more layered than that. Not every quote from a reliable source is allowed to be included. Bobby Lawndale (talk) 11:48, 20 September 2024 (UTC)[reply]
  • Keep it out until there is clear consensus to include (as for me I am on the remove side). The burden on inclusion is on the editor seeking to add it. I looked just now and the content has already been removed (or I missed it when I just looked). WP:BI is not a generally speaking reliable source and thus should not be used for this sort of controversial criticism, WP:NOCRIT and WP:BLPRESTORE apply, do not re-add without consensus. The arguments above that this is not criticism is laughable, its not a notable person making this statement, nor is it a notable publication, it is simply a junk website writing a hit piece and we do not include such junk on wikipedia, there is already plenty of policy on this. If you still disagree, run an RFC (I suspect it will fail miserably, but you never know...) Thanks! Jtbobwaysf (talk) 20:57, 22 March 2024 (UTC)[reply]
    "The arguments above that this is not criticism is laughable" (what?) and "it is simply a junk website writing a hit piece and we do not include such junk on wikipedia" are WP:TENDENTIOUS disputes over the reliability of a source which was deemed reliable on the noticeboard (see WP:SOURCEGOODFAITH). Zenomonoz (talk) 23:21, 22 March 2024 (UTC)[reply]
We've already had this discussion, see Wikipedia:Reliable_sources/Noticeboard/Archive_423#Business_Insider_on_Lex_Fridman there's consensus among established editors that BI is fine to use in this case. Hemiauchenia (talk) 21:00, 22 March 2024 (UTC)[reply]
The quote from the computational biologist discussed there seems fine to me, particularly because it's attributed to Lior Pachter. So I agree with the consensus in that case, but I stand by my objection to the inclusion of the quote I brought up here attributed to "research experts". Chase Kanipe (talk) 23:13, 22 March 2024 (UTC)[reply]
I don't think you, nor Jtbobwaysf understand how "consensus" works on Wikipedia. It does not work by vote. It works by strength of the arguments. Considering the source itself has already been deemed reliable by the noticeboard, there is already a consensus for inclusion based on experienced editors assessments of policy. Zenomonoz (talk) 23:18, 22 March 2024 (UTC)[reply]
I think that's a bit uncharitable reading of Chase's comment. I still don't see a consensus to remove the quote though. Hemiauchenia (talk) 23:24, 22 March 2024 (UTC)[reply]
Looks like you are replying to a few separate comments. Zenomonoz (talk) 23:27, 22 March 2024 (UTC)[reply]
Fixed. Hemiauchenia (talk) 23:31, 22 March 2024 (UTC)[reply]
This response (1) fails to engage with my comment, (2) contradicts your earlier statements, and then (3) repeats the same misunderstanding of my argument I attempted to correct earlier.
  1. I never said anything about or related to voting. I never even referred to "consensus" to bolster my case - I only made reference to the RFC that Hemiauchenia and yourself also referred to as a consensus.
  2. In your first comment to me you dismissed my argument saying Wikipedia is not for me to "make up your own analysis or argument". Now you're saying consensus is about the strength of the arguments. If so then I look forward to you engaging directly with the arguments I've made for why my preferred editorial decisions better conform with Wikipedia policy.
  3. You say there is "already a consensus for inclusion". I will say it again: I have not once suggested that the Business Insider article shouldn't be used as a source as was decided in the RFC. I have only argued against the inclusion of one particular sentence that mixes a statement of fact with a personal insult. There is not a consensus about this particular sentence.
Chase Kanipe (talk) 02:49, 23 March 2024 (UTC)[reply]
I fail to see any good reasoning here beyond your own "I don't like it". Nothing in BLP or NPOV guidelines suggests anything of the sort. It's from a source deemed reliable. You are misunderstanding neutrality to mean "50/50" which is not the point of the policy. Zenomonoz (talk) 04:27, 23 March 2024 (UTC)[reply]
Again here you are projecting arguments onto me I never made and in this case it also appears you are also misattributing up quote? I've never made an argument on the basis of "50/50" and I never said the line you quote as "I don't like it", nor have I made a similar argument.
Please read my arguments again. The essence of my objection is that this sentence mixes a statement of fact with a personal insult against Fridman. I have argued it violates WP:BLP by labeling a person with a contentious label (as un-serious) when WP:BLP requires someone be commonly referred to that way in reliable sources. With this sentence the paragraph also fails to read as conservative or disinterested. Beside's that it personally insults Fridman, I have also argued that this case differs from the previously discussed quote because it is sourced from anonymous "research experts". It also violates the WP:NPOV suggestion that reputation comments are appropriate for inclusion when they're widespread (this opinion is only expressed in a single source), and it also qualifies as a disparaging statement that's "vague" (what does "not taken seriously" mean?). Chase Kanipe (talk) 11:59, 23 March 2024 (UTC)[reply]
The short discussion that was linked to seems to be not clear the hurdle set by WP:LOCALCONSENSUS and as such isn't really much more than we are having again here. WP:BI is a dubious source and as such as yellow in RSP. There is a note over at RSP that there was a time period from 2021-2023 that this was an RS, when it was called only "insider" but this source link clearly states BI. Maybe this is more like a some type of opinion post written by Julia Black and we could consider over at WP:RSN if this is due for inclusion on this article, or do an RFC here on it as well. I think a good question is if this type of opinion piece is an RS and if such content clears the hurdle of WP:NOCRIT. Or is this simply a colleague of the article subject with an axe to grind. Its unfortunate that we have these discussions again and again here at WP, and about inclusion of this type of defamatory content on articles, where the source is often a blog post or an opinion source. X person says Y about Z person and thus we spend all this energy talking about it here at wikipedia (when is anyone else talking about it? Has NYT discussed this source and found this opinion piece interesting?) In my opinion, there is certainly no consensus that this is an RS for this type of CRIT content and as such its odd we are even having this discussion. Maybe we should delete this reception section entirely, the real question is if any of this is even due, let alone the opinion of a BI writer, which is obviously UNDUE. If we do feel that Julia's opinion of Lex is due, why not just state that in the article directly? I for one, dont think any of it is due. Jtbobwaysf (talk) 06:00, 23 March 2024 (UTC)[reply]
No Jtbobwaysf. It isn't a "local consensus", this particular BI article was discussed on the reliable sources noticeboard (Wikipedia:Reliable sources/Noticeboard/Archive 423#Business Insider on Lex Fridman) where it was determined acceptable by four independent users. It is nothing like the unreliable content the RSP aims to deal with.
Refusing to accept the conclusions of independent input from the noticeboard is a WP:NORFC comment, so it's best to stop making this argument. As for WP:CRIT, that is a Wikipedia editor's essay, it is not a real editing guideline. Nothing in the essay would even support removing all reliable source commentary on Lex's career, given that essay actually recommends having a "reception" section. Zenomonoz (talk) 06:19, 23 March 2024 (UTC)[reply]
Apologies, I didnt realize that was from RSN, I had thought it was a brief discussion on this talk page and now archived. Since RSN thinks it is kosher, then obviously local consensus doesnt apply and my above comments in relation to this article/content are moot. I retract my above comments about this content. I still think these commentary reaction sections are not encyclopedic, but that is a larger topic and not suitable for this talk page. Thanks! Jtbobwaysf (talk) 07:49, 23 March 2024 (UTC)[reply]
Lex is far better known for his podcasting than his AI research, and it's not clear he actually does any active research at MIT currently, given the lack of publications over the last several years. AI is also a very high citation field, which makes the 2000+ citations somewhat less impressive in that context. Hemiauchenia (talk) 23:31, 22 March 2024 (UTC)[reply]
In the Boston Globe article it states that he's doing research he hopes to publish in 2024 and he's still employed as a research scientist. So I think he's at least actively attempting to do research. You're right his research record is quite limited especially for a 40 year old, but I think if the article's going to note that it needs to be phrased in a way that isn't a personal attack. I'm not sure how notable any information about his research is besides the autopilot debacle anyway. Chase Kanipe (talk) 02:22, 23 March 2024 (UTC)[reply]
I agree pretty much entirely, it's nonspecific and vague and uses loaded language. Basically undue. SmolBrane (talk) 18:54, 25 March 2024 (UTC)[reply]
To reiterate, I am no longer arguing anything. I was confused about the link and thought it was an archived small discussion of this article's talk page (eg local consensus), until a user thankfully made it obvious to me that the consensus had been formed at RSN, not locally. RSN is exactly the forum for this sort of thing and we should follow that. I apologize for any confusion. Thanks! Jtbobwaysf (talk) 05:59, 27 March 2024 (UTC)[reply]

Edit warring

[edit]

FeldmarschallGneisenau – can you stop edit warring on this?

  • This is not an appropriate source for a birth name. It's a random document that lacks any context. The paragraph stating an alleged birth name is submitted by the journalist Julia Black. This is not the same as the government confirming his birth name.
  • Throwing in statements like "It is unclear when precisely the name change occurred or why" is your own editorialising.
  • Inserting that he up in the "Tajik part of the Soviet Union", is not in the source cited.
  • "In 1994" is not in the source.

Basically, WP:STICKTOSOURCE and stop using a completely inappropriate primary source. Zenomonoz (talk) 11:06, 14 April 2024 (UTC)[reply]

Also note that the three-revert rule is about to be, or has been, passed. —Panamitsu (talk) 11:08, 14 April 2024 (UTC)[reply]
Never mind, they look to be clear sock. Have reported. Zenomonoz (talk) 11:35, 14 April 2024 (UTC)[reply]

Information about Fridman's Background at MIT

[edit]

Lex Fridman's background in artificial intelligence is important information that ought to be placed somewhere very clearly on this page, including the lead. Fridman is best known at MIT for the being main figure involved in putting together many of MIT's online deep learning materials, which include videos of classes he taught at MIT on deep learning, including several courses on autonomous vehicles (6.S094). These materials also include many other videos of guest lecturers teaching at his seminar, which Fridman organized and which eventually became his Artificial Intelligence podcast. Here are many videos of all that:

Fridman also organized additional educational materials about deep learning, GANs, and so on, and made them publically available. For instance, here are some Jupyter notebooks and tutorials, and an article walking through how to use them with TensorFlow, put together by Lex at MIT:

Here's a pretty exhaustive set of videos both from Fridman and guest lecturers from a user on Reddit, with about 60 hours of video total:

Fridman still works as a researcher at MIT's Laboratory for Information and Decision Systems, and is officially affiliated with them in that capacity:

As a researcher, he's listed either as lead or co-author on about 70 papers, typically on computer vision and with a focus on autonomous vehicles. These papers have gathered about 3000 citations overall and his most recent paper was published last year.

This is all very basic, well-established information about Lex. It is all well-sourced, common knowledge and not a secret. However, most of this isn't even on the page at all. Much of it used to be on this page but was removed somewhere around here:

As a result of the political squabbling from last year, the page in its current form is so lopsided that things barely make sense. The majority of the MIT section is now about criticism of one of his papers. There isn't even a link to deeplearning.mit.edu anywhere, which is an omission so egregious as to be completely misleading. These things used to be on the page, but have been scrubbed. Instead, the majority of the references on the page are now from one article called "Peace, love, and Hitler: How Lex Fridman's podcast became a safe space for the anti-woke tech elite".

This page has clearly fallen apart after the political edit warring from last year and needs to be entirely restructured. Criticism is certainly important and should be included - as should support - but neither of these things should dominate every section of the article to such an extent that deeplearning.mit.edu is removed.

The main piece of information to know about this guy's academic background is that he's an MIT AI researcher and lecturer who became a podcaster. At MIT, he was the main public face of their online deep learning lectures for several years. One of the papers he was lead author on was criticized, but other papers he authored were received very well (i.e. DeepTraffic). He was super active at MIT for several years teaching and organizing guest lectures, which eventually evolved into his podcast. He's since refocused on the podcast but is still affiliated at MIT. That sums it up. 2600:1002:B0C6:1D79:F17A:4A45:1963:89FC (talk) 07:35, 9 May 2024 (UTC)[reply]

We dont use reddit, but for sure we can use MIT primary to show that he has lectures on this subject. Jtbobwaysf (talk) 08:51, 9 May 2024 (UTC)[reply]
That would be totally fine to put in external links, but that’s about it. Zenomonoz (talk) 09:01, 9 May 2024 (UTC)[reply]
The IP comment misunderstands what Wikipedia is for, and how it is written. It’s based on secondary independent sources. It isn’t a CV where we list every lecture or paper people wrote. Whats covered in secondary sources is WP:DUE for inclusion. The rest, not so much. Personal objections to what the sources cover is irrelevant. The article reflects what is written in secondary sources. The link to MIT lectures are fine in external links, but that’s it. Zenomonoz (talk) 09:08, 9 May 2024 (UTC)[reply]
This statement "The main piece of information to know about this guy's academic background is that he's an MIT AI researcher and lecturer who became a podcaster" (the users WP:OR) is probably true and we need an WP:RS for this. We cant just infer this from the MIT website. Jtbobwaysf (talk) 10:30, 9 May 2024 (UTC)[reply]