colinhayhurst.bsky.social
CEO @mojeek | No-Tracking Search Engine | Start-up addict | Own views |
https://bio.link/colinhayhurst
41 posts
97 followers
104 following
Regular Contributor
Active Commenter
comment in response to
post
Thank you, good to know and thanks for clarifying. It's what they sent out in their email to us @mojeek.com and presumably many others.
comment in response to
post
What good does it do us when we respect robots.txt, publish our IPs, and do not donate to any such; nor would if we could. As for the Google-Reddit deal, I can't now say more than this: www.404media.co/google-is-th...
comment in response to
post
This is available now: ofcomlive.my.salesforce-sites.com/formentry/Re...
comment in response to
post
Congratulations
comment in response to
post
Pull the plug
comment in response to
post
The access to hyperlinks increases the risk of users discovering uncomfortable truths
comment in response to
post
Pots & Kettles
comment in response to
post
No source but see this. www.microsoft.com/en-us/startu...
They were definitely using outside sources and as main (can't/shouldn't say which). Note we have a search API @mojeek so i woudl know ;)
comment in response to
post
Almost certainly not. Perplexity will have switched. Bing racked up their prices and restricted usage of Bing API with LLMs in mid-2023. Brave didn't have a search API then but have now.
comment in response to
post
"Request No. 7 seeks documents sufficient to show the types & quantity of data used to train certain models that OpenAI has developed or licensed for use in connection with any
potential or actual Generative AI Tool or AI Search Tool."
😂 that will reveal what GPT-4 was trained on.
Et toi Google? 🤡
comment in response to
post
You might argue it's now too easy to "make something people want". For example, small biz want leads and we now have more and more lead gen pimpers; selling data that is likely to be crap and unethically collected.
Let's "make something people value"
comment in response to
post
🤣
comment in response to
post
Unverified Propaganda as a Service.
comment in response to
post
Alphabet cash on hand for Q3 was $93B anyway
comment in response to
post
agreed
comment in response to
post
and very strong data sharing provisions for qualifying competitors
comment in response to
post
Indeed they don't
comment in response to
post
Is that what is meant by "term" in "Google may not release any other Google Browser during the term of this Final
Judgment absent approval by the Court."? So "(10) years from the Effective Date"
comment in response to
post
Very large non-EU technology firms have dominated our market by scaling fast; with $$$ to attract talent and land-grab
comment in response to
post
useful, thanks
comment in response to
post
No (public) comment
comment in response to
post
Not to mention the problem of linking generated content to original. Probabilistic and partial Spotify for AI; think again folks.
comment in response to
post
"for real change to occur one needs to convince other people who are not yet sufficiently aware about privacy and surveillance, or those who feel they are too busy to bother."
My colleague Josh @mojeek.bsky.social provides tips on how bring others on board:
blog.mojeek.com/2024/10/priv...
comment in response to
post
Well yes, but Duck search is Bing on Azure, and Kagi on GCP using Google and Mojeek APIs. So to be independent of US infra you need to use mojeek.com or API.
comment in response to
post
Thanks for tip. Just done same.
comment in response to
post
Good initiative! I've been thinking of extending www.w3.org/2022/tdmrep/ for this kind of application. One thing to keep in mind is that the problem predates LLMs (though it was impossible to get any eyes on it prior to that explosion). For instance, voice assistants are a problem too.
comment in response to
post
@technollama.bsky.social @robin.berjon.com
comment in response to
post
"that they were not aware" seems more plausible to me 🤐 or even a third option that they and lawyers both knew.
or fourth: "because all you need is to purchase one copy of a book" - visions of big tech acquiring libraries