MGM's IT team just rolled craps

Welcome to Runtime! Today: a massive cyberattack has thrown one of the biggest casino operators in the U.S. into disarray, how Elon Musk's Christmas data-center adventure caused all kinds of problems for Twitter, and the latest funding rounds in enterprise tech.

(Was this email forwarded to you? Sign up here to get Runtime each week.)

Nice casino you got there

MGM Resorts remained unable to provide many basic computerized services well into Tuesday after it was knocked offline Monday by what it called "a cybersecurity issue" that left slot machines dark and guests unable to get into their own rooms.

The FBI said it was investigating after MGM said it shut down several systems in hopes of containing the damage, which security experts believe suggests some sort of ransomware attack. The fallout from the incident went nationwide across MGM's properties but was particularly evident in Las Vegas.

• The Las Vegas Review-Journal reported that guests were unable to check into their hotel rooms or even access their hotel rooms with digital keys at various times on Monday.
• The incident also shut down slot machines across the country, and the Review-Journal estimated that 60% of the slot machines at the MGM Grand were still down on Tuesday morning.
• The company was also unable to take sports bets across the country ahead of the thrilling Jets-Bills Monday Night Football game, the first one of the season.
• MGM Resorts makes about $43 million a day across its entire operation, judging by last quarter's results, and this incident will leave a mark.

MGM remained tight-lipped about the cause of the incident Tuesday.

• It's impossible to know right now what may have led to the issue, especially after a year in which vulnerabilities in an unheralded file-transfer application caused one of the biggest cybersecurity incidents in a long time.
• We do know that MGM Resorts runs a hybrid cloud infrastructure and uses AWS as a cloud provider.
• The company also used Oracle's Fusion Cloud ERP to centralize and modernize its billing systems, which appeared to be one of the main services affected by the incident.
• And it's been a few years since this was announced, but the company also appeared to be running Microsoft's Office 365.

If this incident is indeed a ransomware attack, pressure will likely grow inside the company to pay the ransom and get its systems back before it loses too much more revenue.

• Security experts really don't want victims of ransomware attacks to pay the ransom, because it just encourages the entire criminal enterprise to keep at it.
• However, unless this is a weird remake of Ocean's Eleven using a cybersecurity incident as a distraction to rob the vault, one thing MGM has a lot of is cash money.
• It took a casino operator in Canada two weeks to recover from a ransomware incident earlier this year, and MGM is a much larger operation that is likely (hopefully) scrambling to enable its backup and recovery systems.

Public companies like MGM are required to disclose any ransomware payments within 24 hours, and provide notice of material breaches of their cybersecurity infrastructure within four days, according to new rules enacted by the SEC earlier this year.

Elon vs the data center

Two days before Christmas last year, Elon Musk crawled underneath a server rack in one of Twitter's data centers in Sacramento. He used a pocket knife "to jimmy open an electrical cabinet, pulled the server plugs, and waited to see what happened. Nothing exploded."

That was Musk's plan for moving thousands of servers from Sacramento to the Portland suburbs last December, dismissing the concerns of Twitter employees who warned him that such a rushed move could backfire, according to CNBC's excerpt of Walter Isaacson's biography of Musk released this week. In order to save money to service the debt he saddled the company with upon purchasing it last year, Musk wanted to be out of the Sacramento data center as soon as possible and deputized a coalition of the willing over the objections of his infrastructure team.

Musk, his cousins, and every moving truck they could find in the Sacramento area got the job done, but at great disruption to Twitter's server infrastructure. "For the next two months, X was destabilized. The lack of servers caused meltdowns, including when Musk hosted a Twitter Spaces for presidential candidate Ron DeSantis," according to the excerpt, and Musk would later admit the whole thing was a mistake.

Enterprise funding

Imbue landed $200 million in Series B funding (!) that values the AI foundational-model research company at over $1 billion.

D-Matrix raised $110 million in Series B funding to further the development of chips designed for AI inference.

Certa scored $35 million in Series B funding to build out its compliance and risk-management software portfolio.

The Runtime roundup

Oracle's stock fell the most it had ever suffered in a single day over the last 20 years after it reported earnings that missed expectations and provided a cautious outlook for the rest of the year.

Salesforce CEO Marc Benioff talked about AI a lot while surrounded by a bunch of fake plastic trees.

Nvidia plans to release new software next month that it claims could double the performance of the coveted H100 GPU.

Microsoft used 34% more water across its data centers last year, which environmental experts believe is tied to its rapid buildout of AI servers.

Every Patch Tuesday is followed closely, but CISA put special emphasis on today's release of bug fixes that patch two vulnerabilities that are being actively exploited.

Thanks for reading — see you Thursday!