Bing Blogs

Thank you for providing this.

Bing Chat runs on GPT-4, however [1]. And Bing gets this wrong in all 3 of its modes (Creative, Balanced, and Precise) as of time of writing.

Given this experiment and similar others presented around here, it stands to reason that GPTs(**1) often identify(**2) the problem as a "wolf, goat, and cabbage" problem and then merely guess which node of the problem is the middle node (inner node of the "danger to" graph), yielding a 1/3 chance of getting it right by pure luck, resulting in diverse reports here.

(**2) That does not always yield an adequate response beyond the mere permutation of nodes, however. I've been getting the following variants for step 1. from Bing in Precise in response to marginally slightly different rewordings of the same:

- The general escorts the ambassador of Costaguana through the tunnel first. This leaves the ambassador of Atlantis and the ambassador of Buranda in the bunker, but they are not alone because the general is still there.

- The general escorts the ambassador of Costaguana through the tunnel first. This leaves the ambassador of Atlantis and the ambassador of Buranda in the bunker, but they are not alone because they have each other.

and so on.

(**1) I also tried Bard and Llama 2 with even more disastrous results full of nonsense of (**2) kind. The earlier posted response of ChatGPT-3.5 is also prime with these as well.

Re

> By the way, as soon as these systems are able to check their reasoning (i…

Please read a bit of the history of this debate:

https://stackoverflow.com/questions/36702789/google-crawler-for-single-page-application

Till today the Google Crawler is the only one that gets even near good search results of SPAs. That's because they have to run a full browser to load every page to parse it. Every. Single. Page. That's why it is hard to do correctly. Read a bit about their current approach here:

https://developers.google.com/search/docs/crawling-indexing/javascript/javascript-seo-basics#what-the-user-sees-what-the-crawler-sees

Specially:

> Keep in mind that server-side or pre-rendering is still a great idea because it makes your website faster for users and crawlers, and not all bots can run JavaScript.

Recently in 2019 Bing Bot just gave up and is using the same approach a Google:

https://blogs.bing.com/webmaster/october-2019/The-new-evergreen-Bingbot-simplifying-SEO-by-leveraging-Microsoft-Edge

So nowadays having a SPA isn't as bad for SEO as it used to be, but again, and I'll keep nailing on this, that is only because we spent the entire last decade throwing millions of dollars and hundreds of hours of engineer time to make SPAs a viable option.

> In fact, I'm pretty sure any competitive search engine these has to have a very complex crawler that is more than able to deal with javascript rendered pages.

Save the region-specific engines which likely lag behind, Google [0] and Bing [1] both support crawling javascript, and Bing is generally the search engine index of choice for all other search engines like Yahoo, DDG (at least for now, I occasionally get crawls from duckduckgobot), etc.

0: https://developers.google.com/search/docs/guides/javascript-seo-basics#how-googlebot-processes-javascript

1: https://blogs.bing.com/webmaster/october-2018/bingbot-Series-JavaScript,-Dynamic-Rendering,-and-Cloaking-Oh-My

More direct AMP link: https://amp.reddit.com/r/explainlikeimfive/comments/5d4x2h/eli5_the_purposebenefit_of_googles_new_amp/

The original page is hosted on the original domain. What other hosts provide are AMP caches to display them faster. For example, Microsoft Bing:

- https://blogs.bing.com/search/September-2016/bing-app-joins-the-amp-open-source-effort

- https://blogs.bing.com/Webmaster-Blog/September-2018/Introducing-Bing-AMP-viewer-and-Bing-AMP-cache

- https://www.bing-amp.com/c/s/amp.reddit.com/r/explainlikeimfive/comments/5d4x2h/eli5_the_purposebenefit_of_googles_new_amp/

We do have an internal Google at my place of work. It is called "Bing for business" and i will assume that many companies are using that too.

Here is a old article about what is

https://blogs.bing.com/search/2017-09/finding-what-you-need-at-work-just-got-easier-with-bing-for-business

Edit: I am not the sysadm and Bing for Business might have changed name since the article i linked to was written but we definitely have a "internal Google" that makes information across all our internal properties searchable when using bing.com

Training data is an issue. Microsoft released some datasets of buildings in North America (125 million footprints). Later, they decided to use the same technology to generate building footprints in Uganda and Tanzania:

https://blogs.bing.com/maps/2019-09/microsoft-releases-18M-building-footprints-in-uganda-and-tanzania-to-enable-ai-assisted-mapping

They needed different training data to cope with the different landscapes.

Facebook has faced similar challenges trying to use ML to generate road centerlines for OpenStreetMap.