DOJ Goes after Google for Advertising Monopoly Power

DOJ seeks breakup of Google ad unit, but it's like an old song that we know all too well.

Hey Guys,

The U.S. Department of Justice and eight states have filed an antitrust lawsuit against Google, calling for a breakup of the search-giant's advertising technology business.

I’ve been thinking about this all day. The way it’s covered is just a flash in the pan too.

The DOJ going after Google is old news, but there’s always a new slant. New York media always gives you the sense that this is serious and potentially Earth-shaking but it never ends up being true. I have more faith in EU regulation or even India, than I do the U.S. taking rule of law seriously in corporate affairs.

The Justice Department is seeking the breakup of Google’s business brokering digital advertising across much of the internet, a major expansion of the legal challenges the company faces to its business in the U.S. and abroad (WSJ).

Unfortunately Silicon Valley’s power as some rather poor outcome for real innovation in A.I., and A.I. ethics, alignment and regulation in general.

The sun never sets on Google's antitrust woes

But Silicon Valley never truly faces any kind of legal justice around its monopoly power. The consequences on American innovation are worse than the media reports. Of late, IPO activity isn’t keeping up with Mergers and acquisitions. The result, BigTech can acquire who and what they want, pretty much without restriction.

The last splurge of Microsoft to the tune of $13 Billion on OpenAI is a case in point. The DOJ’s Google lawsuit is just symbolic of a much bigger problem which spills over into A.I. innovation, or frankly the lack of it.

A Chatbot like ChatGPT that can make stuff up very well, isn’t exactly perhaps a breakthough. For BigTech’s antitrust we have to look at Europe or India for real results. India’s antitrust regulator has fined Alphabet Inc.’s Google 9.4 billion rupees ($113 million) for “abusing” its dominant position on its Android mobile app store, the second such penalty for the tech giant within days.

But the big questions regarding Search Advertising is will Conversational AI be another layer that makes it more or less competitive? Perhaps this will result in Google and Microsoft owning an even bigger share of Search, as if that were even possible?

Economists from Yale and the University of Maryland reveal a dramatic shift in the IPO market that might be hurting consumers.

The economic stagflation of 2022 and 2023 is creating a warped future for startups, IPOs in relation to mergers and acquisitions and faster consolidation as startups run out of money with vastly lower valuations.

DOJ seeks breakup of Google ad unit

If America had wanted to break up its winner-takes-all Capitalism, it would have done so around 2010, and we wouldn’t be in such a stale and advertising laden internet. The latest noise is not to be trusted. Silicon Valley has washington in its back pocket.

Our flashy A.I. might not actually be so great though. Stanford University researchers found that programmers who used OpenAI’s Codex, a model that generates computer code, were more likely to produce buggy software than those who coded from scratch. Here we are talking about Microsoft’s GitHub Copilot. I believe ChatGPT coding has likely spiked malware and phishing to even higher levels in 2023.

Google’s anti-competitive strategy in Advertising is not something new, it’s built on black-hat practices on many many levels. One part of Google’s strategy, the complaint alleges, was to acquire other companies to grow its power in the advertising market and “set the stage for Google’s later exclusionary conduct across the ad tech industry.” Those acquisitions included a 2008 purchase of publisher ad server DoubleClick and a “nascent ad exchange” that would become Google’s AdX. This allowed Google to require publishers in some instances to use all of its tools to gain access to any one, rather than working with rival tools for parts of the online ad-buying process.

So why isn’t the DOJ looking into Microsoft throwing money at OpenAI? Or will it have to wait until 2033 or 2035, after the damage has already been done? Innovation does require a level playing field and an incentive for actual top researchers to become entrepreneurs and niches were true competition is actually possible.

ChatGPT owner OpenAI projects $1 billion in revenue by 2024

ChatGPT expects $Billions in profits from ChatGPT and GPT-4 like integrations, so where is the anti-trust when you need it?

It’s hard to trust the DOJ or the State lawsuits against BigTech that has become just so powerful in terms of lawyers and lobbying. The department’s earlier lawsuit (CNBC), filed in October 2020 under the Trump administration, accused Google of using its alleged monopoly power to cut off competition for internet search through exclusionary agreements. That case is expected to go to trial in September.

Now I’m not an anti-monopoly lawyer but it seems the DOJ has a rather poor track record against Facebook, Google, Apple and the like. Microsoft’s Activision and OpenAI deal are fairly nefarious if you ask me for fair competition to take place. They bottleneck Microsoft’s ability to dominate gaming and some aspects of consumer A.I. in a totally unfair way to leverage Microsoft’s own Advertising ambitions.

Increasingly Apple, Amazon and Microsoft want a share of the Advertising duopoly Google and Facebook have moated with incredible lack of innovation. The 2023 to 2026 is pivotal for the future of advertising and in some ways “A.I. hype” is just the Nursemaid. All the while we have influencers and journalists talking as if A.I. was a real and legit thing.

Consumer A.I. companies like ByteDance leverage apps like TikTok. Something which Microsoft, Google or even Facebook (without acquisitions) would have nearly no way or idea how to build. Our biggest players are simply not very innovative or innovation product centric. Microsoft acquiring LinkedIn in retrospect is anti-competitive in pretty nefarious ways.

The public perceives OpenAI's ChatGPT as revolutionary, but the same techniques are being used and the same kind of work is going on at many research labs, says the deep learning pioneer, Yann LeCun. A.I. innovation isn’t what you think, if you tune out the hype a clearer picture begins to emerge:

Yann LeCun @ylecun

LLMs are still making sh*t up. That's fine if you use them as writing assistants. Not good as question answerers, search engines, etc. RLHF merely mitigates the most frequent mistakes without actually fixing the problem.

Delip Rao @deliprao

the success of chatgpt has lead to investors thinking RLHF is magic (to some extent it is), but boy they are going to be disappointed when their portfolios realize its limitations

A lot of hype on LinkedIn or Twitter amounts to misinformation to be honest. Substack and beehiiv I have noticed are inheriting a lot of the hype I saw this firsthand with a bunch of Subtack and beehiiv Newsletters on the topic of GPT-4. I found it all pretty disheartening to say the least. The truth is we don’t know a ton about GPT-4 yet, until it’s released or announced likely in May, 2023.

Google is set to announce several A.I. products in 2023, several of which we know about thanks to a NYT article recently. So that the DOJ keeps going after Google, decades after its crimes have already been committed and Silicon Valley has warped in Surveillance Capitalism, is not just tragic, it shows a real decline of Silicon Valley standards in tech, regulation and ultimately, in A.I. as well.

The Internet is Fu*king Broken