Chinese AI Lab Deepseek Grinds Ahead…Allegedly
December 31, 2024
Is the world’s most innovative AI company a low-profile Chinese startup? ChinaTalk examines “Deepseek: The Quiet Giant Leading China’s AI Race.” The Chinese-tech news site shares an annotated translation of a rare interview with DeepSeek CEO Liang Wenfeng. The journalists note the firm’s latest R1 model just outperformed OpenAI’s o1. In their introduction to the July interview, they write:
“Before Deepseek, CEO Liang Wenfeng’s main venture was High-Flyer, a top 4 Chinese quantitative hedge fund last valued at $8 billion. Deepseek is fully funded by High-Flyer and has no plans to fundraise. It focuses on building foundational technology rather than commercial applications and has committed to open sourcing all of its models. It has also singlehandedly kicked off price wars in China by charging very affordable API rates. Despite this, Deepseek can afford to stay in the scaling game: with access to High-Flyer’s compute clusters, Dylan Patel’s best guess is they have upwards of ‘50k Hopper GPUs,’ orders of magnitude more compute power than the 10k A100s they cop to publicly. Deepseek’s strategy is grounded in their ambition to build AGI. Unlike previous spins on the theme, Deepseek’s mission statement does not mention safety, competition, or stakes for humanity, but only ‘unraveling the mystery of AGI with curiosity’. Accordingly, the lab has been laser-focused on research into potentially game-changing architectural and algorithmic innovations.”
For example, we learn:
“They proposed a novel MLA (multi-head latent attention) architecture that reduces memory usage to 5-13% of the commonly used MHA architecture. Additionally, their original DeepSeekMoESparse structure minimized computational costs, ultimately leading to reduced overall costs.”
Those in Silicon Valley are well aware of this “mysterious force from the East,” with several AI head honchos heaping praise on the firm. The interview is split into five parts. The first examines the large-model price war set off by Deepseek’s V2 release. Next, Wenfeng describes how an emphasis on innovation over imitation sets his firm apart but, in part three, notes that more money does not always lead to more innovation. Part four takes a look at the talent behind DeepSeek’s work, and in part five the CEO looks to the future. Interested readers should check out the full interview. Headquartered in Hangzhou, China, the young firm was founded in 2023.
Cynthia Murrell, December 31, 2024
OpenAI Partners with Defense Startup Anduril to Bring AI to US Military
December 27, 2024
No smart software involved. Just a dinobaby’s work.
We learn from the Independent that “OpenAI Announces Weapons Company Partnership to Provide AI Tech to Military.” The partnership with Anduril represents an about-face for OpenAI. This will excite some people, scare others, and lead to remakes of the “Terminator.” Beyond Search thinks that automated smart death machines are so trendy. China also seems enthused. We learn:
“‘ChatGPT-maker OpenAI and high-tech defense startup Anduril Industries will collaborate to develop artificial intelligence-inflected technologies for military applications, the companies announced. ‘U.S. and allied forces face a rapidly evolving set of aerial threats from both emerging unmanned systems and legacy manned platforms that can wreak havoc, damage infrastructure and take lives,’ the companies wrote in a Wednesday statement. ‘The Anduril and OpenAI strategic partnership will focus on improving the nation’s counter-unmanned aircraft systems (CUAS) and their ability to detect, assess and respond to potentially lethal aerial threats in real-time.’ The companies framed the alliance as a way to secure American technical supremacy during a ‘pivotal moment’ in the AI race against China. They did not disclose financial terms.”
Of course not. Tech companies were once wary of embracing military contracts, but it seems those days are over. Why now? The article observes:
“The deals also highlight the increasing nexus between conservative politics, big tech, and military technology. Palmer Lucky, co-founder of Anduril, was an early, vocal supporter of Donald Trump in the tech world, and is close with Elon Musk. … Vice-president-elect JD Vance, meanwhile, is a protege of investor Peter Thiel, who co-founded Palantir, another of the companies involved in military AI.”
“Involved” is putting it lightly. And as readers may have heard, Musk appears to be best buds with the president elect. He is also at the head of the new Department of Government Efficiency, which sounds like a federal agency but is not. Yet. The commission is expected to strongly influence how the next administration spends our money. Will they adhere to multinational guidelines on military use of AI? Do PayPal alums have any hand in this type of deal?
Cynthia Murrell, December 27, 2024
The Future: State Control of Social Media Access, Some Hope
December 25, 2024
It’s great that parents are concerned for their children’s welfare, especially when there are clear and documented dangers. The Internet has been in concerned parents’ crosshairs since its proliferation. Back in the AOL days it was easier to monitor kids access, you simply didn’t allow them to log on and you reviewed their browser history. However, with the advent of mobile devices and the necessity of the Internet for everyday living, parents are baffled on how to control their children and so is the Australian government. In an extreme case, the Australian parents proposed a bill to ban kids under the age of sixteen from using social media. The Senior relates how they are winning the battle: “Parents To Lose Final Say In Social Media Ban For Kids.”
The proposed bill is from Prime Minister Anthony Albanese’s administration and it plans to ban all kids under the age of sixteen from any and other social media platforms. Parents are taken out of the equation entirely. Parents will not be allowed to consent and many see it as a violation of their civil and parental rights.
The bill hasn’t been drafted yet and probably won’t be in 2024. It is believed that the first legislation on the bill will be in 2025 and will slowly work its way through the Australian parliament. The blanket ban would also not require age verification:
“Asked if parents would be allowed to consent to their children being on social media at a younger age, Communications Minister Michelle Rowland told Labor’s party room meeting “no”. She said people using social media would not have to upload proof of identity directly to those platforms, when minimum age requirements kick in. ‘The opposition is the only party arguing that people should upload 100 points of ID and give it to TikTok,’ she told the meeting. The government wants 12 months of consultation to figure out exactly how the ban will be enforced.”
Australia doesn’t have faith in parents’ efforts to regulate their kids on social media, so the government is acting in the kids’ best interests. It does sound like the government is overstepping, but social media experts and mental health professionals have documented the potential and real harm of social media on kids. Many parents also don’t monitor and discipline their children’s Internet usage habits. Is this an overstep by the government? No, just a first step.
Whitney Grace, December 25, 2024
VoIP in Russia, Nyet. Telegram Voice, Nyet. Just Not Yet
December 24, 2024
Written by a dinobaby, not an over-achieving, unexplainable AI system.
PCNews.ru in everyone’s favorite special operations center reported that Roskomnadzor (a blend of the FBC and a couple of US three letter agencies) has a New Year’s surprise coming. (Xmas in Russia is often celebrated on January 7, 2024.) The short write up reported to me in English via the still semi reliable Google Translate that calls within “messenger apps” are often fraudulent. I am not sure this is a correct rendering of the Russian word. One of my colleagues suggested that this is a way to say, “Easily intercepted and logged by Roskomnadzor professionals.”
Among the main points in the article are:
- The target is voice traffic not routed via Roskomnadzor
- Communication operators — that is, Internet providers, data centers, etc. — are likely to be required to block such traffic with endpoints in Russia
- As part of the “blocks,” Roskomnadzor wants to identify or have identified for the entity such functions as “identifying illegal call centers (including those using SIM boxes) on their networks.”
The purpose is to protect Russian “consumers.” The source cited above included an interesting factoid. YouTube traffic, which has been subject to “blocks” has experienced a decrease in traffic of 80 percent.
Not bad but a 20 percent flow illustrates that Roskomnadzor has been unable to achieve its total ban. I wonder if this 80 percent is “good enough” for Roskomnadzor and for the fearless head of state who dictates in Russia.
Stephen E Arnold, December 24, 2024
FOGINT: Intelware Tension Ticks Up
December 24, 2024
Observations from the FOGINT research team.
On Friday, December 20, 2024, NSO Group, the Pegasus specialized software outfit, found itself losing a court squabble with Facebook (Meta and WhatsApp). According to the Reuters’ news story pushed out at 915 pm Eastern time, “US Judge Finds Israel’s NSO Group Liable for Hacking in WhatsApp Lawsuit.” In case you don’t have the judgment at hand, you can find the United States District Court, Norther District of California document at this link.
The main idea behind the case is that the NSO Group’s specialized software pressed into duty for the purpose of obtaining information about WhatsApp users. The mechanism was to exploit “a bug in the messaging app to install spy software allowing unauthorized surveillance.” NSO Group’s fancy legal two step did not work.
The NSO Group has become the poster child for the “compromise the mobile” phone and obtain data. The Pegasus system exfiltrates data and, when properly configured, can capture information from a mobile device. Furthermore, the company’s hassles about its customers’ use of the Pegasus tool unwittingly created a surge in software and specialized services performing identical or similar tasks.
The FOGINT team has identified firms which have found different ways of compromising mobile devices. The company, therefore, has been an innovator and its approach to compromising devices has [a] focused attention on Israel’s technical competence in this specialized software niche and [b] rightly or wrongly illustrated that the technology can act with extreme prejudice when used by some clients to solve what they perceive as “problems.”
There are several larger consequences which the FOGINT team has identified:
- Specialized software is more prevalent because the revelations about Pegasus have encouraged entrepreneurs and technologists to develop more effective surveillance methods
- Unique delivery methods have been crafted. These range for in-app malware to more sophisticated multi-stage malware installed as a consequence of a user’s carelessness
- Making clear that powerful surveillance tools can be installed in a way that does not require the user to click, email, or interact. The malware simply dials up a mobile and bingo! the device is compromised.
How will this judgment affect the specialized software industry? In FOGINT’s view, the decision will further stimulate competition and the follow of novel surveillance techniques. One consequence also may be that law enforcement and intelligence professionals will encounter headwinds when similar specialized software is required for certain investigations. FOGINT’s view is that NSO Group’s go-go approach to sales created a problem for the company and for specialized software. Some technologies should remain “secret,” which is now becoming an old-fashioned viewpoint. Marketing is not always a benefit.
Stephen E Arnold, December 24, 2024
Thales CortAIx (Get It?) and Smart Drones
December 23, 2024
Countries are investing in AI to amp up their militaries, including naval forces. Aviation Defense Universe explores how one tech company is shaping the future of maritime strategy and defense: “From Drone Swarms To Cybersecurity: Thales’ Strategic AI Innovations Unveiled.” Euronaval is one of the world’s largest naval defense exhibitions and CortAlx Labs at Thales shared their innovations AI-power technology.
Christophe Meyer is the CTO of CortAlx Labs at Thales and he was interviewed for the above article. He spoke about the developments, innovations, and challenges his company faces with AI integration in maritime and military systems. He explained that Thales has three main AI divisions. He leads the R&D department with 150 experts that are developing how to implement AI into system architectures and cybersecurity. The CortAlx Labs Factory has around 100 hundred people that are working to accelerate AI integration into produce lines. CortAlx Lab Sensors has 400 workers integrating AI algorithms into equipment such as actuators and sensors.
At Euronavel Thales, Meyer’s company demonstrated how AI plays a crucial role in information processing. AI is used in radar operations and highlights important information from the sensors. AI algorithms are also used in electronic warfare to enhance an operator’s situation awareness and pointing out information that needs attention.
Drones are also a new technology Thales is exploring. Meyer said:
“Swarm drones represent a significant leap in autonomous operations. The challenge lies in providing a level of autonomy to these drones, especially when communication with the operator is lost. AI helps drones in the swarm adapt, reorganize, and continue their mission even if some units are compromised. This technology is platform-agnostic, meaning it applies to aerial, maritime, and terrestrial swarms, with the underlying algorithms remaining consistent across domains.”
Drones are already being used by China and Dubai for aerial shows. They form pictures in the night sky and are amazing to watch. Ukraine and Russia are busy droning one another. Exciting.
Whitney Grace, December 23, 2024
The EU Cafeteria Wants to Serve Grilled Google
December 10, 2024
This write up was created by an actual 80-year-old dinobaby. If there is art, assume that smart software was involved. Just a tip.
How does one cook a kraken (a Norwegian octopus)? Here’s the recipe from Garlic & Zest:
- Clean the octopus, remove the beak(s) and place them in a large, heavy pot or Dutch oven.
- Add the vegetables, wine and corks.
- Bring to a boil, reduce heat to a simmer and cook for 45 minutes to one hour.
The hapless octopus awaits its fate. The goal is to serve up tasty individual dishes and follow up with a refreshing takoyaki. Thanks, MidJourney. Looks tasty.
I want to point out that the creature dies in this process. Now to the write up:
“Google Split Still on the Table, New EU Antitrust Chief Says” reports:
A potential split of Google’s business is still under consideration, according to Teresa Ribera, the European Union’s new competition chief, who also pledged to build bridges with incoming US President Donald Trump.
That’s the intent to grill the delectable sea monster, according to some children’s books.
The person setting the menu and supervising the chefs who will chop off the tentacles, remove its beak (ouch!), tenderize the helpless creature, and plop it on the barbie is Teresa Ribera.
For those who don’t follow Spain’s emergent leaders, Ms. Ribera is a socialist who will find some philosophical points of difference between her new kitchen team and the Wild West chuckwagon approach taken toward Google in the US of A.
The cited news story says the new EU Antitrust chef (sorry, I meant chief) allegedly said:
“It’s [chopping up Google] something that is of course on the table, and we try to work together with other relevant competition authorities worldwide, including the US competition authorities,” she said. ‘It is important to take into consideration this potential division, divestment of some of these businesses. We will be assessing case-by-case.”
The question is, “When will the main course be served?” Restaurant kitchens — like Brussels, the French and German governments — can be chaotic places.
Stephen E Arnold, December 10, 2024
Amazon Offers to Fight Crime, Not on Its Platform But in the District of Columbia No Less
December 10, 2024
This blog is created by a dinobaby and his helpers who are neither AI nor dinobabies. If there is fancy art, please, assume that smart software contributed. Dinobabies can barely think let alone draw.
Amazon has been busy explaining how its smart software will make Alexa Live Again! Meanwhile some legal eagles have been dropping documents on the digital bookstore and its happy, happy employees and contractors. This news reached me via the for-sale outfit CNBC, a talking heads program on “real” TV. “Amazon Sued by DC Attorney General for Allegedly Excluding Neighborhoods from Prime Delivery” reports what may be obvious to anyone who has worked in the District of Columbia or gone for a late night walk alone on a hot summer evening in some interesting parts of the District. Oh, the slogan for the DC entity is “Taxation without Representation.”
Senior professionals for a high-technology company reach consensus. The giant firm will help law enforcement address certain issues. Thanks, MidJourney. Good enough.
The CNBC story says:
Washington, D.C.’s attorney general sued Amazon on Wednesday [yep, CNBC, that was December 4, 2024], accusing the company of covertly depriving residents in certain ZIP codes in the nation’s capital from access to Prime’s high-speed delivery. The lawsuit from AG Brian Schwalb alleges that, since 2022, Amazon has “secretly excluded” two “historically underserved” D.C. ZIP codes from its expedited delivery service while charging Prime members living there the full subscription price. Amazon’s Prime membership program costs $139 a year and includes perks like two-day shipping and access to streaming content.
The idea is that a DC resident pays $139 a year to get “Prime” treatment only to get the shaft; that is, no delivery for you, pilgrim.
Amazon took time out from its Alexa Lives Again! activities to issue a statement. According to the “real” news source a really happy Amazon professional allegedly said:
Amazon spokesperson Kelly Nantel said in a statement it’s “categorically false” that its business practices are “discriminatory or deceptive.” “We want to be able to deliver as fast as we possibly can to every zip code across the country, however, at the same time we must put the safety of delivery drivers first,” Nantel said in a statement. “In the zip codes in question, there have been specific and targeted acts against drivers delivering Amazon packages. We made the deliberate choice to adjust our operations, including delivery routes and times, for the sole reason of protecting the safety of drivers.” Nantel said Amazon has offered to work with the AG’s office on efforts “to reduce crime and improve safety in these areas.”
I like the idea of a high technology outfit trying to reduce crime. I have been the victim of Amazon fraud. I like to mention the women’s underwear shipped to me instead of the $600 Ryzen CPU I ordered. A more recent example was Amazon’s emailing me a picture of a stranger’s door with the message, “Your package has been delivered.” I spent about an hour trying to get a human to address the issue. The response, I think, was we will credit you for the order. When? How? Sorry, I have no idea because the happy Amazon professional based in a third party customer service facility had no clue. I am thinking, “Why doesn’t Amazon do something about the fraud on its own platform?”
I personally have zero confidence that Amazon can address such issues in the District of Columbia. It is known to one resident — my son, who lives in the District — has observed porch pirates following Amazon delivery vans. As soon as the package is left, the porch pirate exits the chase vehicle and takes the package. The fix ranges from designated pick up points in certain vulnerable areas of the District to having a security team follow the porch pirates on their rental scooters, Teslas, and other vehicles. Of course, the alternative is to go to a store like REI in the District of Columbia, shop, and return home. (Oh, sorry, I forgot that Amazon’s business practices have contributed to the demise of brick and mortar retail. Oh, well. It was just a dinobaby thought.)
This will be an interesting legal case to follow. Porch pirates follow Amazon delivery vehicles and prove the value of paying attention.
Stephen E Arnold, December 10, 2024
Google and 2025: AI Scurrying and Lawsuits. Lots of Lawsuits
December 6, 2024
This is the work of a dinobaby. Smart software helps me with art, but the actual writing? Just me and my keyboard.
I think there are 193 nations which are members of the UN. Two entities which one can count but are what one might call specialty equipment organizations: The Holy See aka Vatican City and the State of Palestine. The other 193 are “recognized,” mostly pay their UN dues, and have legal systems of varying quality and diligence.
I read “Google Earns Fresh Competition Scrutiny from Two Nations on a Single Day.” The write said:
In India – the most populous nation on Earth – the Competition Commission ordered [PDF] a probe after a developer called WinZo – which promotes itself with the chance to “Play Mobile Games & Win Cash” – complained that Google Play won’t host games that offer real money as prizes, only allowing sideloading onto Android devices.
Then it added:
Advertising is the reason for the other Google probe announced Thursday, by the Competition Bureau of Canada – the world’s second-largest country by area. The Bureau announced its investigations found Google’s ads biz “abused its dominant position through conduct intended to ensure that it would maintain and entrench its market power” and “engaged in conduct that reduces the competitiveness of rival ad tech tools and the likelihood of new entrants in the market.” The Bureau thinks the situation can be addressed if Google sells two of its ads tools – but the filing in which the identity of those two products will be revealed is yet to appear on the site of the Competition Tribunal.
Whether Google is good or evil is, in my opinion, irrelevant. With the US, the EU, Canada, and India chasing Google for its alleged misbehavior, other nations are going to pay attention.
Does that mean that another 100 or more nations will launch their own investigations and initiate legal action related to the lovable Google’s approach to business? In practical terms what does this mean?
- Google will be hiring lawyers and retaining firms. This is definitely good for legal eagles.
- Google will win some, delay some, and lose some cases. The losses, however, will result in consequences. Some of these will require Google to write checks for penalties. These can add up.
- Conflicting decisions are likely to result in delays. Those delays means that Google will be more Googley. The number of ads in YouTube will increase. The mysterious revenue payments will become more quirky. Commissions on various user-customer-Google touch points will increase.
Net net: We have a good example of what a failure to regulate high technology companies for a couple of decades creates. Kicking the can down the road has done what exactly?
Stephen E Arnold, December 6, 2024
China Seeks to Curb Algorithmic Influence and Manipulation
December 5, 2024
Someone is finally taking decisive action against unhealthy recommendation algorithms, AI-driven price optimization, and exploitative gig-work systems. That someone is China. ”China Sets Deadline for Big Tech to Clear Algorithm Issues, Close ‘Echo Chambers’,” reports the South China Morning Post. Ah, the efficiency of a repressive regime. Writer Hayley Wong informs us:
‘Tech operators in China have been given a deadline to rectify issues with recommendation algorithms, as authorities move to revise cybersecurity regulations in place since 2021. A three-month campaign to address ‘typical issues with algorithms’ on online platforms was launched on Sunday, according to a notice from the Communist Party’s commission for cyberspace affairs, the Ministry of Industry and Information Technology, and other relevant departments. The campaign, which will last until February 14, marks the latest effort to curb the influence of Big Tech companies in shaping online views and opinions through algorithms – the technology behind the recommendation functions of most apps and websites. System providers should avoid recommendation algorithms that create ‘echo chambers’ and induce addiction, allow manipulation of trending items, or exploit gig workers’ rights, the notice said.
They should also crack down on unfair pricing and discounts targeting different demographics, ensure ‘healthy content’ for elderly and children, and impose a robust ‘algorithm review mechanism and data security management system’.”
Tech firms operating within China are also ordered to conduct internal investigations and improve algorithms’ security capabilities by the end of the year. What happens if firms fail? Reeducation? A visit to the death van? Or an opportunity to herd sheep in a really nice area near Xian? The brief write-up does not specify.
We think there may be a footnote to the new policy; for instance, “Use algos to advance our policies.”
Cynthia Murrell, December 5, 2024
-
- Subscribe to Beyond Search
Feature archive
News archive
-
