Cyber Scene - Flying High: Blue Skies or Blue Screen?

Image removed.

By krahal

The Washington Post's Adela Suliman, Dan Rosenzweig-Ziff, Eva Dou and María Luisa Paúl reported that the CrowdStrikd outage, identified on 18 July, affected not only 4,700 canceled flights and 37,600 delayed (8,800 in the US) but also websites, broadcasters and payment systems.

Microsoft explained that "…the 'preliminary root cause' appeared to be 'a configuration change in a portion of our Azure backend workloads.'" This led to "…interruption between storage and compute resources which resulted in connectivity failures that affected downstream Microsoft 365 services dependent on these connections." It continued, sketching out that in addition to travelers, many other entities were not so lucky either. Hospitals in the US, the UK, Germany, and Israel, inter alia, were also impacted. Britain's National Health Service had to cancel surgeries on 19 July and resorted to paper patient records (!). Sloan Kettering Cancer Center on 19 July canceled "…any procedure that requires anesthesia." Emergency phones were down temporarily on Friday from Alaska to Ohio. Fortunately, the Paris Olympics, with the opening ceremony this week, were not victims. On the other hand, US governmental entities were not completely spared: Social Security Administration and the Department of Education were hit-- a technical, not political issue.

The very day, 18 July, of Microsoft's outage onset, the Economist reports that a cybersecurity startup, Wiz, created by four former Israeli soldiers in 2015, was now worth $12B and its creators were in talks with Google's Alphabet to acquire Wiz. Interestingly, these four sold their first startup to Microsoft where they worked until 2020. Their approach is to collect large numbers of data so that customers can "…spot vulnerabilities that hackers could exploit." To help reshape cybersecurity technology, Wiz is moving, "inexorably," into the cloud. Secondly, it is also moving toward one-stop shops to make it easier for users to manage and use tools to talk to each other. "Being swallowed by Alphabet gives Wiz a giant platform atop the world's third biggest cloud." The article continues, itemizing Alphabet's recent purchases leading to its $2 trillion in 2020 annual revenue. On 23 July, however, Wiz reportedly called off talks with Alphabet, deciding to focus on an Initial Public Offering.

AI continues to get above-the-fold tech attention. Some, however, still believe that there has been little economic impact on AI technology. The 2 July Economist, is one such critic. The article acknowledges that the Big Five (Alphabet, Amazon, Apple, Meta and Microsoft) are believers in AI's future.

The author, however, cites the slow rate of AI adoption as one of the problems. Despite the usage, the statistics of less than 5% of businesses using AI in the last 2 weeks are quite low. Canada and the UK are in similar situations. As companies begin to use AI, they acknowledge setbacks. The article cites problems with deciding where to invest, or being unable to jump in, as AI develops so fast, which leads to out-of-date status in an unacceptable short time.

The study does accept the use of various companies in various AI applications, but the factor of lost jobs in the US, for example, is not happening now; the article cites that white collar employment is a percentage higher than before the pandemic. In fact, the article states: "Some economists think AI will transform the global economy without booting people out of jobs. Collaboration with a virtual assistant may improve performance." A paper from experts of the University of Chicago and Copenhagen University arrived at the conclusion that among 100,000 workers, ChatGPT halved the time spent on one-third of work tasks, "…in theory a big boost to efficiency…but little evidence of a surge in productivity." What is projected, however, is that big tech users of AI may create a bonanza returning high share prices of users as well as the providers of AI, but that might not arrive until after 2032. The "revolution" is expected to surge very slowly.

There are counter examples that underscore a much sooner success.

The first is the US Department of Defense. Defense One's Edward Graham cites the DoD's AI Task Force Lima, launched in August 2023, to help officials craft guidelines and identify infrastructure needs for emerging capabilities, as reported by the Pentagon's Chief Digital and Artificial Intelligence Office (CDAO), Radha Plumb. She explained that this allows for "…the understanding of guidelines and guardrails we need to put in place." The objective is to "allow DOD personnel to undertake 'responsible risk-taking' within clearly defined parameters."

The Washington Post's (the Post) Alexander C. Karp and Nicholas W. Zamiska look at AI in the larger world of future weapons "…that will eclipse the atom bomb."

The two--one the co-founder and CEO of Palantir Technologies and the second head of its corporate affairs and legal counsel--look at what John Lewis Gaddis calls the "long peace" as far as world wars go. The world will be different. General Mark Milley (ret.) recent Chairman of the Joint Chiefs of Staff asks: "Do we really think a manned aircraft is going to be winning the skies in 2088?" Of course not. However, the guest writers concede: "But with the rise of artificial intelligence and the use of large language models to make targeting recommendations on the battlefield, the relationship is shifting. Now software is at the helm, with hardware--the drones in Ukraine and elsewhere--increasingly serving as the means by which the recommendations of AI are carried out."

They continue, explaining that the young Americans who know how to build AI systems "…are often also the most ambivalent about working for the military." They see Silicon Valley as detached from the understanding that the peace they enjoy is due to the US military. Guest writers continue: "We should, as a society, be capable of carrying on a debate about the merits of using military force abroad without hesitation to provide those sent into harm's way with the software they need to do their jobs." They close by saying the important question is not what new generation of autonomous weapons incorporating AI will be, but who will build them.

The Wired's Sharon Lemac-Vincere offers an example of commercial technology pairing up with military needs to prevent cyberattacks "above our heads." While this occurs not with boots on the ground but rather with satellites and other hardware in space, foreign cyberattacks could be restrained. This partnership would include not only spy or military communications satellites, but also commercial spacecraft as well. All need protection.

DOD has created a new partnership to advance this: Commercial Augmentation Space Reserve (CASR). CASR would "…enhance US national security and the country's competitive advantage in space." CASR underscores that many countries have advanced space capability with their commercial sector, rapidly advancing beyond government's capabilities. The article goes on to note win-win opportunities and eliminating redundancy in using off-the-shelf technology when possible.

On a more personal level, the New York Times' (the Times) Brian X. Chen addresses the possibility that AI smartphones and computers will require users to provide more data. Given one's daily spam collection, many users may be reluctant to provide more data. It could be viewed as intrusive:

In this new paradigm, your Windows computer will take a screenshot of everything you do every few seconds. An iPhone will stitch together information across many apps you use. And an Android phone can listen to a call in real time to alert you to a scam.

There are concerns about the lack of "a vote" to what is shared. Since snapshots would be taken every five seconds, they are to be stored and analyzed directly on the PC, not reviewed by Microsoft, or used to improve AI. Security researchers are not comfortable, as the data "…could easily expose everything you've ever typed or viewed if it was hacked." Microsoft decided to postpone the release indefinitely. 

To see previous articles, please visit the Cyber Scene Archive.

Submitted by grigby1 CPVI on