Something I've been musing about recently (and semi work related): I think we're probably in the tail-end period of usability and value of the public internet. With the proliferation of content generated by LLM's which are trained on the internet data, but also trained to generate content which "sounds believable" our of the aggregate data (without any actual knowledge or ability to check validity), the reliability and usefulness of the public internet data will decline over time. As the bot-garbage propagates, it will likely overwhelm "real" information, even without people leveraging LLM's for disinformation (as is already happening, with successful effect, because people are relatively easy to influence). I wonder if, at some point, we will be telling stories to our grand children about the period of time when the information which was free on the internet was also reasonably good and reliable, and not a wasteland of bot-garbage and propaganda spewed into the world, to be consumed by perpetually ill-informed people, and/or those conditioned to be easily entertained by zero-value content (not unlike what most of cable television has become now).
Nick Carroll’s Post
More Relevant Posts
-
Warm greetings to all Tech Enthusiasts! We are back with another blog, titled "The Intricate Difference Between Sections 67 and 67A of of the IT Act, 2000. Obsence content is available and accessible through the internet with little known repercussions for publishing the same under Section 67 of the IT Act, 2000. Obscenity has been a debated question since a significant amount of time which is visible through the rulings of the landmark case of Regina v. Hecklin. In the Indian context, the article ventures into demarcating the difference between Section 67 and 67A of the IT Act. Here, the author finds Section 67A, to be more lexically objective than Section 67. On a concluding note the author cautions about the loose language used under Section 67 which can call for a cause of action under irrelevant situations, namely, defamation. Link: https://lnkd.in/dReuuEFA
To view or add a comment, sign in
-
The Danish DPA's "Chromebook case" may have far-reaching consequences for both public and private organisations🧐 The case is interesting from several perspectives, in particular because it involves protecting fundamental rights of children and because it has lead to claims of changes in national law by several organisations. But my primary interest lies with the legal analysis conducted⚖️, stripped from political opinions. And when we zoom in on that🕵🏼♂️, it seems that the case has wide consequences for both public and private organsiations using BigTech - and basically any tech - data processors. So, do you want to avoid a "Chromebook-gate" of your own and learn more? Check out this blog post and sign up for the webinar to hear the elaborated story and practical guideline⬇️ Link to the full case from the Danish DPA and link to Danish blogpost are in the comments. https://lnkd.in/dhkbkijh
Google Chromebook in Public Schools | The Case That Became a Saga
complycloud.com
To view or add a comment, sign in
-
Very well thought-out discussion of the shifts in "free speech" due to the online nature of many of our communications. This sentence in particular stood out: "We need a robust public conversation about what we, the people, want this space to look like, and what practices and guardrails are likely to strengthen the ties that bind us in common purpose as a democracy." My immediate response: "Absolutely. But ... where would we have such a public conversation?" Many have proposed social media itself as the new public square, but (even if we disregard the impact of algorithmically-displayed feeds) participants on even the most widespread social media platform represent just a small fraction of the people and perspectives in the world. Geographically, we're also often still separated, even as the Internet might connect us ideologically. What do YOU see as the modern equivalent of the public square? https://lnkd.in/gW32TMT5
The One Internet Hack That Could Save Everything
wired.com
To view or add a comment, sign in
-
President @ OnPoint Data Strategy | Navigating Success with OnPoint Precision | The Data Behind Your ROI
Interesting article highlighting the impact that PURPOSEFULLY programed bias in a LLM can have. While I understand the attempt to avoid bias in a dataset, this is what can happen when you purposefully try to correct for that by programming additional bias / counter-bias into the model. It begs the question, how do we address transparency with LLM's to identify what other programed biases also exist within their models? ... "Earlier this week, a former Google employee posted on X that it’s “embarrassingly hard to get Google Gemini to acknowledge that white people exist,” showing a series of queries like “generate a picture of a Swedish woman” or “generate a picture of an American woman.” The results appeared to overwhelmingly or exclusively show AI-generated people of color." https://lnkd.in/e6HtsRjx
To view or add a comment, sign in
-
-
It’s hard to know what’s true: The result of the fragmented information landscape and reality of human cognitive wiring is that it’s hard to know what’s true. While there’s no shortage of information today about any given topic, finding the truth amid the overflow is much harder than it should be. Google search brings up relevant results but doesn’t really help you separate fact from fiction. As an example, consider the Covid-19 lab leak theory. It consists of a whirlpool of contradicting reports (even from reputable sources and different parts of the US government), intricate details, and ever-changing information. Figuring out what to believe is quite challenging, requiring the skills and effort of an investigative journalist, which most of us don’t have the time and energy to invest in. Recognizing this complexity, Dialectic proposes a transformative solution.
To view or add a comment, sign in
-
Digital Markets, Consumers and Competition Bill: with rumours that the PM has been lobbied to allow wider grounds of appeal from decisions of the Competition and Markets Authority imposing (or decisions not to impose) conduct requirements on big tech, the JWP of UK Bars and Law Societies has put in a paper - which I drafted with Nigel Parr and others - setting out another proposal: an administrative appeal.
To view or add a comment, sign in
-
Professor of Economics & BB&T Distinguished Professor of Capitalism at Western Carolina University, Executive Director at Public Choice Society, co-author Madmen, Intellectuals, and Academic Scribblers (Stanford Press)
In social science research, non-results are notoriously difficult to publish. Evidently not so with the effects of voter ID laws, however. A swift skimming of Google Scholar turns up lots of articles in some top journals and by top political scientists that essentially amount to a collective "Meh". Not much impact on turnout (suppression) or integrity (perceived fraud), with some exceptions around the strictness of the voter ID laws, and of course a good bit of arguing over measurement and specification issues. Friends who have studied this: What is the deal -- i.e., what effects does the balance of the evidence point to? Substantiation, not snark, please. Asking for about forty friends.
To view or add a comment, sign in
-
https://lnkd.in/dqM4yNW6 Availability bias is the tendency to think easily accessible information is the most factual information.
'The Google effect' can lead you into this common mental trap—here's how to avoid it
cnbc.com
To view or add a comment, sign in
-
This 2012 paper, by Pace University professor Adam Klein, is cited as the first to describe the concept of Information Laundering. Abstract Many studies in recent years have addressed the notable ways that Internet features such as blogs and search engines have democratized the community of information seekers and providers, however, fewer investigations have addressed the darker element that has emerged from that same democratic sphere. That is, the huge resurgence and transformation of racist communities across cyberspace. This article presents a new theory of information laundering to explain the process by which racial hate speech is becoming legitimized through a borrowed network of online associations. This Internet-specific theory builds upon research of “information-based” racist propaganda to explain how today's search engines, social networks, and political blogs unwittingly enable purveyors of bigotry to infiltrate into mainstream spaces of public discourse. https://lnkd.in/edFicEVM
Slipping Racism into the Mainstream: A Theory of Information Laundering
academic.oup.com
To view or add a comment, sign in
-
On Trump and the 14th Amendment... When anti-democratic forces exploit the openness of the democratic system to claim power (what I call "infiltrative insurgency"), it is crucial that the state can *constitutionally* bar that force from the ballot. Yet, implementation will be very delicate. With "militant democracy," the standards of shutting down the offending actor must be clear and the judiciary cannot act in a partisan manner. People will view that last one differently, hence the high stakes involved societally and need to manage perceptions. Easier said than done, particularly in a highly polarized America Another complication is balancing timing and evidence. Moving too early will be to forego evidence of malfeasance, but it stands a higher likelihood of being effective. Moving too late means accruing plenty of evidence, but also a reduced ability to do anything about it. There's a whole literature on this—"militant democracy," also called "intolerant democracy." It is a crucial component of response to "Trojan Horse" actors who understand that the best way to conquer the state is to exploit its own systems and openness. https://lnkd.in/eexnBZ8J. I believe Trump fits this description, and there are many others out there. In the end, this question echoes Popper's tolerance paradox: "If we extend unlimited tolerance even to those who are intolerant, if we are not prepared to defend a tolerant society against the onslaught of the intolerant, then the tolerant will be destroyed, and tolerance with them."
Infiltrative Insurgency the Trojan Horse
academic.oup.com
To view or add a comment, sign in