Aftershocks hit Microsoft, Google and OpenAI
Antitrust inquiries, EU regulation, Gemini launch spoiled, and Ilya in limbo
The dust at OpenAI has begun to settle but the drama continues to generate aftershocks. This week we’ve seen signs that this story is far from over.
Google attempted to seize the advantage, only to flub the launch (again). Meanwhile antitrust authorities on both sides of the Atlantic have stirred from their slumber to take a closer look at Microsoft’s ties to OpenAI.
And OpenAI’s Brutus has vanished.
Google can’t handle the truth
Remember when Google flubbed the launch of Bard a few months ago? An event that was intended to be a victory lap instead turned into an embarrassment.
The same thing happened again this week. During the past two days, the euphoria that attended the big reveal for Google’s largest language model, Gemini, dissipated when news emerged that the best demo was faked. This report tells the sad story.
Google has admitted that the demo was edited to exaggerate the capabilities of Gemini. The question remains: why?
What’s puzzling about this incident is that fakery wasn’t necessary. Gemini is a significant achievement. Google DeepMind has much to be proud of.
So why blow it by exaggerating?
This is more than bad communications or a marketing fail. It seems to be a pattern, and that indicates poor judgment at the leadership level.
It is a shame because there were some important nuggets buried in Google’s announcements. Nano is a big deal. Multimodality is, too. Google has the talent and the market power to compete fiercely against OpenAI.
As I wrote two days ago, what Alphabet lacks is leadership. Blunders like this prove it.
The European Union has reached an agreement on AI regulation
Lawmakers in the EU have reached a deal on terms for the regulation of AI. To become law, the final text must be agreed by member nations, the members of the European Parliament, and the members of the European Commission, which is the executive body of the EU.
A paradox that might become the fly in the ointment: EU commissioner Thierry Bretton declared, “The AIAct is much more than a rulebook—it’s a launchpad for EU start-ups and researchers to lead the global AI race.” That sounds good until you read this paragraph:
European companies have expressed their concern that overly restrictive rules on the technology, which is rapidly evolving and gained traction after the popularisation of OpenAI’s ChatGPT, will hamper innovation. Last June, dozens of some of the largest European companies, such as France’s Airbus and Germany’s Siemens, said the rules were looking too tough to nurture innovation and help local industries.
For months we’ve been hearing grumbling from our associates in Europe that the EU gets regulation and unemployment while the US has no meaningful tech regulation and full employment.
Satya Nadella’s Intervention in the OpenAI drama has drawn the scrutiny of the EU’s antitrust authorities
The European Union’s Competition and Markets Authority (CMA) has opened an inquiry into Microsoft’s relationship with OpenAI with an Invitation to Comment (ITC). From the CMA statement:
The CMA has been closely monitoring the impact of partnerships and strategic agreements which could result in a weakening of competition in the development or use of FMs (foundation models). The partnership between Microsoft and OpenAI (including a multi-year, multi-billion dollar investment, collaboration in technology development and exclusive provision of cloud services by Microsoft to OpenAI) represents a close, multi-faceted relationship between two firms with significant activities in FMs and related markets.
There have recently been a number of developments in the governance of OpenAI, some of which involved Microsoft. In light of these developments, the CMA is now issuing an ITC to determine whether the Microsoft / OpenAI partnership, including recent developments, has resulted in a relevant merger situation and, if so, the potential impact on competition.
The CMA will review whether the partnership has resulted in an acquisition of control – that is, where it results in one party having material influence, de facto control or more than 50% of the voting rights over another entity – or change in the nature of control by one entity over another.
The US Federal Trade Commission has also opened an antitrust inquiry into Microsoft’s relationship with OpenAI
What’s that fishy smell? The FTC is investigating.
According to Bloomberg, Microsoft did not report the OpenAI deal to the Federal Trade Commission because of the firm’s non-profit status.
The whole point of Microsoft taking a minority stake in OpenAI was to avoid antitrust scrutiny. The best laid plans…
Game of Clones: Ilya Sutskever in limbo
On November 17, OpenAI’s chief scientist, Ilya Sutskever, delivered the news to CEO Sam Altman that he was fired. That message started a chain reaction that ultimately caused the CEO of Microsoft to intervene forcefully to restore Sam Altman to the throne lest the company implode.
Now Altman is back in the catbird seat at OpenAI, reporting to a new board of directors that no longer includes Sutskever.
Nobody knows for sure what Ilya’s role will be at OpenAI going forward. He has not been seen in the office this week.
Ilya remains in limbo. This season of Game of Thrones is far from over.
Re: the EU's draft and evolving regulations. Industry says the regulations will stifle innovation. Human rights group say holes in the regulations will endanger lives. It is all standard playbook stuff. The cliche is a good compromise leaves everyone a little unhappy. I just listened to a Harvard panel discussion on Free Speech and Hate Speech. The applicable take-away is we know how to address problems; what the exercises and techniques for finding balance and compromise are. We just rarely see situations where impactful players all both (or all) sides of an argument are motivated to participate in them. What we are seeing on AI in the US right now is compromise-avoidance.