AI

AI isn’t and won’t soon be evil or even smart, but it’s also irreversibly pervasive

Comment

LLM word with icons as vector illustration. AI concept of Large Language Models
Image Credits: Getty Images

Artificial intelligence — or rather, the variety based on large language models we’re currently enthralled with — is already in the autumn of its hype cycle, but unlike crypto, it won’t just disappear into the murky, undignified corners of the internet once its “trend” status fades. Instead, it’s settling into a place where its use is already commonplace, even for purposes for which it’s frankly ill-suited. Doomerism would have you believe that AI will get so smart it’ll enslave or sunset humanity, but the reality is that it’s much more threatening as an omnipresent layer of error and hallucinations that seep into our shared intellectual groundwater.

The doomerism versus e/acc debate continues apace, with all the grounded, fact-based arguments on either side that you can expect from the famously down-to-earth Silicon Valley elites. Key context for any of these figures of influence is to remember that they spend their entire careers lauding/decrying the extreme success or failure of whatever tech they’re betting on or against — only to have said technology usually fizzle well-short of either the perfect or the catastrophic state. Witness everything always, forever, but if you’re looking for specifics, self-driving is a very handy recent one, as is VR and the metaverse.

Utopian versus dystopian debates in tech always do what they’re actually intended to do, which is distract from having real conversations about the real, current-day impact of technology as it’s actually deployed and used. AI has undoubtedly had a massive impact, particularly since the introduction of ChatGPT just over a year ago, but that impact isn’t about whether we’ve unwittingly sown the seeds for a virtual deity, it’s about how ChatGPT proved far more popular, more viral and more sticky than its creators ever thought possible — even while its capabilities actually matched their relatively humble expectations.

Use of generative AI, according to most recent studies, is fairly prevalent and growing, especially among younger users. The leading uses aren’t novelty or fun, per a recent Salesforce study of use over the past year; instead, it’s overwhelmingly being used to automate work-based tasks and communications. With a few rare exceptions like when it’s used for preparing legal arguments, the consequences of some light AI hallucination in generating these communications and corporate drudgery are insignificant, but it’s also undoubtedly resulting in a digital strata that consists of easy-to-miss factual errors and minor inaccuracies.

That’s not to say people are particularly good at disseminating information free of factual error; rather the opposite, actually, as we’ve seen via the rise of the misinformation economy on social networks, particularly in the years leading up to and including the Trump presidency. Even leaving aside malicious agendas and intentional acts, error is just a baked-in part of human belief and communication, and as such has always pervaded shared knowledge pools.

The difference is that LLM-based AI models do so casually, constantly and without self-reflection, and they do so with a sheen of authoritative confidence which users are susceptible to because of many years of relatively stable, factual and reliable Google search results (admittedly, “relatively” is doing a lot of work here). Early on, search results and crowdsourced online pools of information were treated with a healthy dose of critical skepticism, but years or even decades of fairly reliable info delivered by Google search, Wikipedia and the like has short-circuited our distrust of things that come back when we type a query into a text box on the internet.

I think the results of having ChatGPT and its ilk producing a massive volume of content with questionable accuracy for menial everyday communication will be subtle, but they’re worth investigating and potentially mitigating, too. The first step would be examining why people feel like they can entrust so much of this stuff to AI in its current state to begin with; with any widespread task automation, the primary focus of inquiry should probably be on the task, not the automation. Either way, though, the real, impactful big changes that AI brings are already here, and while they don’t look anything like Skynet, they’re more worthy of study than possibilities that rely on techno-optimistic dreams coming true.

More TechCrunch

Tags

Cargo ships docking at a commercial port incur costs called “disbursements” and “port call expenses.” This might be port dues, towage, and pilotage fees. It’s a complex patchwork and all…

Shipping logistics startup Harbor Lab raises $16M Series A led by Atomico

AWS has confirmed its European “sovereign cloud” will go live by the end of 2025, enabling greater data residency for the region.

AWS confirms will launch European ‘sovereign cloud’ in Germany by 2025, plans €7.8B investment over 15 years

Go Digit, an Indian insurance startup, has raised $141 million from investors including Goldman Sachs, ADIA, and Morgan Stanley as part of its IPO.

Indian insurance startup Go Digit raises $141M from anchor investors ahead of IPO

Peakbridge intends to invest in between 16 and 20 companies, investing around $10 million in each company. It has made eight investments so far.

Food VC Peakbridge has new $187M fund to transform future of food, like lab-made cocoa

For over six decades, the nonprofit has been active in the financial services sector.

Accion’s new $152.5M fund will back financial institutions serving small businesses globally

Meta’s newest social network, Threads, is starting its own fact-checking program after piggybacking on Instagram and Facebook’s network for a few months.

Threads finally starts its own fact-checking program

Looking Glass makes trippy-looking mixed-reality screens that make things look 3D without the need of special glasses. Today, it launches a pair of new displays, including a 16-inch mode that…

Looking Glass launches new 3D displays

Replacing Sutskever is Jakub Pachocki, OpenAI’s director of research.

Ilya Sutskever, OpenAI co-founder and longtime chief scientist, departs

Intuitive Machines made history when it became the first private company to land a spacecraft on the moon, so it makes sense to adapt that tech for Mars.

Intuitive Machines wants to help NASA return samples from Mars

As Google revamps itself for the AI era, offering AI overviews within its search results, the company is introducing a new way to filter for just text-based links. With the…

Google adds ‘Web’ search filter for showing old-school text links as AI rolls out

Blue Origin’s New Shepard rocket will take a crew to suborbital space for the first time in nearly two years later this month, the company announced on Tuesday.  The NS-25…

Blue Origin to resume crewed New Shepard launches on May 19

This will enable developers to use the on-device model to power their own AI features.

Google is building its Gemini Nano AI model into Chrome on the desktop

It ran 110 minutes, but Google managed to reference AI a whopping 121 times during Google I/O 2024 (by its own count). CEO Sundar Pichai referenced the figure to wrap…

Google mentioned ‘AI’ 120+ times during its I/O keynote

Firebase Genkit is an open source framework that enables developers to quickly build AI into new and existing applications.

Google launches Firebase Genkit, a new open source framework for building AI-powered apps

In the coming months, Google says it will open up the Gemini Nano model to more developers.

Patreon and Grammarly are already experimenting with Gemini Nano, says Google

As part of the update, Reddit also launched a dedicated AMA tab within the web post composer.

Reddit introduces new tools for ‘Ask Me Anything,’ its Q&A feature

Here are quick hits of the biggest news from the keynote as they are announced.

Google I/O 2024: Here’s everything Google just announced

LearnLM is already powering features across Google products, including in YouTube, Google’s Gemini apps, Google Search and Google Classroom.

LearnLM is Google’s new family of AI models for education

The official launch comes almost a year after YouTube began experimenting with AI-generated quizzes on its mobile app. 

Google is bringing AI-generated quizzes to academic videos on YouTube

Around 550 employees across autonomous vehicle company Motional have been laid off, according to information taken from WARN notice filings and sources at the company.  Earlier this week, TechCrunch reported…

Motional cut about 550 employees, around 40%, in recent restructuring, sources say

The keynote kicks off at 10 a.m. PT on Tuesday and will offer glimpses into the latest versions of Android, Wear OS and Android TV.

Google I/O 2024: Watch all of the AI, Android reveals

Google Play has a new discovery feature for apps, new ways to acquire users, updates to Play Points, and other enhancements to developer-facing tools.

Google Play preps a new full-screen app discovery feature and adds more developer tools

Soon, Android users will be able to drag and drop AI-generated images directly into their Gmail, Google Messages and other apps.

Gemini on Android becomes more capable and works with Gmail, Messages, YouTube and more

Veo can capture different visual and cinematic styles, including shots of landscapes and timelapses, and make edits and adjustments to already-generated footage.

Google Veo, a serious swing at AI-generated video, debuts at Google I/O 2024

In addition to the body of the emails themselves, the feature will also be able to analyze attachments, like PDFs.

Gemini comes to Gmail to summarize, draft emails, and more

The summaries are created based on Gemini’s analysis of insights from Google Maps’ community of more than 300 million contributors.

Google is bringing Gemini capabilities to Google Maps Platform

Google says that over 100,000 developers already tried the service.

Project IDX, Google’s next-gen IDE, is now in open beta

The system effectively listens for “conversation patterns commonly associated with scams” in-real time. 

Google will use Gemini to detect scams during calls

The standard Gemma models were only available in 2 billion and 7 billion parameter versions, making this quite a step up.

Google announces Gemma 2, a 27B-parameter version of its open model, launching in June

This is a great example of a company using generative AI to open its software to more users.

Google TalkBack will use Gemini to describe images for blind people