Workflow
Internet
icon
Search documents
快手-W(01024):强烈谴责黑灰产的违法犯罪行为,已向公安机关报警并向相关部门报告
智通财经网· 2025-12-23 04:22
智通财经APP讯,快手-W(01024)发布公告,其快手应用的直播功能于2025年12月22日22:00左右遭到网 路攻击,公司已第一时间启动应预案,经全力处置与系统修复,快手应用的直播功能已逐步恢复正常服 务。快手应用的其他服务未受影响。公司始终严守合规底线,坚决反对任何违规内容及行为。公司强烈 谴责黑灰产的违法犯罪行为,已就上述事宜向公安机关报警并向相关部门报告,并将视情况采取其他适 当的法律补救措施,以保障公司及其股东的权益。 ...
AI狼人杀终极决战!GPT、Qwen、DeepSeek大乱斗,人类高玩汗流浃背
量子位· 2025-12-23 04:16
鹭羽 发自 凹非寺 量子位 | 公众号 QbitAI 我真栓Q了!围观了场 狼人杀 ,看得我汗流浃背…… 半小时全程高能,根本停不下来: 天崩开局倒钩狼悍跳预言家、冲锋狼死于话多、神职上大分每晚都是平安夜。 结果你跟我说,这些玩家都是 AI ??? 果然会玩还得看 淘宝 ~最近他们整活的这个AI狼人杀大乱斗 WhoisSpy.ai ,大模型在里面简直咔咔乱杀。 D老师、Qwen、Kimi、GLM一个个都化身心机boy推拉博弈,be like: …… 不过u1s1,虽然这些Agent看似性格迥异,实则一个个都是狼人杀高玩来着。 而且门槛也不高,自己就能手搓一个出来。 是不是有点手痒了? (咳咳) 不卖关子了,这就是我最近刷到的一个AI狼人杀比赛,还是淘宝办的——首届 「高校生VS开发者对抗赛」 。 展开来说,就是淘宝发了个召集令,广邀高校学生和AI开发者,带着自家Agent来真刀实枪碰一场,看看谁的Agent思维更缜密、更会盘逻 辑。 六边形战士 Kimi :武力值MAX,第六感Next Level。 老实人 DeepSeek :虽然我只是一介平民,虽然我只会划水,但我相信跟对人走对路,奥利给! 喜剧人 Qwe ...
拿走200多万奖金的AI人才,到底给出了什么样的技术方案?
机器之心· 2025-12-23 04:15
Core Viewpoint - The article emphasizes the significant opportunities for young individuals proficient in AI technology in China, particularly highlighted by the recent Tencent Advertising Algorithm Competition, which showcased innovative solutions to complex advertising challenges [2][5]. Group 1: Competition Overview - The Tencent Advertising Algorithm Competition revealed that all top 10 teams received job offers from Tencent, with the champion team awarded a prize of 2 million yuan [2]. - The competition focused on a real-world problem in advertising that lacks a definitive solution, pushing participants to explore practical and innovative approaches [4][5]. Group 2: Advertising Challenges - Advertising is often viewed negatively, but it is essential for the sustainability of many services and content, leading platforms to seek smarter, less intrusive advertising methods [7]. - The competition addressed how to make advertising more targeted and relevant, reducing unnecessary exposure to users [7][16]. Group 3: Methodologies in Advertising - Two primary methodologies in advertising recommendation systems were discussed: traditional discriminative methods and emerging generative methods [8]. - Discriminative methods focus on matching user profiles with ads based on predefined features, while generative methods analyze user behavior over time to predict future interactions [9][14]. Group 4: Competition Challenges - Participants faced challenges related to the scale of data, involving millions of ads and users, while having limited computational resources [21]. - The complexity of the data structure, including multimodal historical behavior data, added to the difficulty of modeling user interactions effectively [21][22]. Group 5: Champion Team Solutions - The champion team, Echoch, introduced a three-tier session system, periodic encoding, and time difference bucketing to enhance the model's understanding of user behavior over time [28][29]. - They developed a unified model capable of switching strategies between predicting clicks and conversions, addressing the differing objectives of these actions [34][36]. - The team also incorporated randomness in ad encoding to improve exposure for less popular ads, significantly increasing their training focus [37]. Group 6: Runner-Up Team Solutions - The runner-up team, leejt, tackled the challenge of handling large-scale data by compressing the vocabulary size and using shared embeddings for low-frequency ads [42]. - They implemented session segmentation and heterogeneous temporal graphs to manage the complexity of user behavior data effectively [44]. - The team optimized engineering processes to maximize GPU utilization, achieving significant performance improvements in model training [48]. Group 7: Industry Implications - The competition highlighted the transition from discriminative to generative models in advertising, with Tencent already implementing generative models in its internal systems, yielding positive results reflected in financial data [51]. - Tencent plans to open-source the competition data to foster community development and explore the potential of real-time personalized advertising generation in the future [52].
微信辟谣
第一财经· 2025-12-23 04:13
请大家不要轻易点击可疑链接,若不小心点击了,切记不要输入或透露个人密码与手机验证码,切实 提高安全意识。 来源: " 谣言过滤器 "微信公号 12月23日,微信官方辟谣平台" 谣言过滤器 " 微信公号发布消息,全文如下: 12月22日晚某直播平台遭到黑灰产攻击,部分传言称 " 违规直播间中隐藏着病毒链接,许多用户点 入后,微信账号即被盗取,不法分子随即向账号好友发送借款请求,实施诈骗。 " 经核实,上述信息不属实。 微信账号有严格的安全保护机制, 截至目前,我们没有发现相关问题和 收到类似反馈。 ...
快手深夜“失守”
财联社· 2025-12-23 03:06
Core Viewpoint - The incident involving Kuaishou highlights the vulnerabilities in content moderation and the growing threat of black and gray market activities in the internet industry [1][2]. Group 1: Incident Overview - On the night of the 22nd, Kuaishou experienced a surge of inappropriate content in multiple live streams, leading to widespread user concern and panic [1]. - Kuaishou attributed the incident to organized attacks from black and gray market groups, which demonstrated advanced technical capabilities to bypass platform defenses [1]. - The event raised questions about Kuaishou's content moderation effectiveness, especially following their recent announcement of AI-driven content governance [1]. Group 2: Platform Responsibilities - As an operator of the online space, Kuaishou must take primary responsibility for content safety, revealing shortcomings in their preemptive defenses and emergency response mechanisms [2]. - The platform needs to enhance its technical capabilities and develop a comprehensive defense network to address evolving threats from black and gray market activities [2]. Group 3: Industry-Wide Implications - The incident serves as a wake-up call for the entire internet industry, emphasizing the need for collaborative efforts among platforms, regulatory bodies, and the public to combat black and gray market activities [3][4]. - The complexity of black and gray market operations necessitates improved inter-departmental cooperation and information sharing to effectively disrupt these criminal networks [4]. - The industry must move away from isolated approaches and establish collaborative risk management mechanisms to enhance overall resilience against attacks [4]. Group 4: Legal and Regulatory Framework - Recent laws and regulations, such as the Cybersecurity Law and the Governance of Online Information Content, have been implemented to combat online chaos, but the intricate nature of black and gray markets poses significant challenges [4]. - Strengthening legal frameworks and increasing penalties for black and gray market activities are essential to deter criminal behavior and protect users [4]. Group 5: Call to Action - The incident underscores the need for a collective effort to safeguard the online ecosystem, with the public encouraged to report violations and platforms urged to reflect on their responsibilities [5]. - Continuous collaboration among all stakeholders is crucial to diminish the operational space of black and gray markets and ensure a safe online environment [5].
快手平台突遭黑灰产攻击 奇安信安全专家汪列军:规模化攻击超出人工审核的应对极限
Xin Lang Ke Ji· 2025-12-23 01:21
Group 1 - Kuaishou experienced a severe attack from hacker organizations, leading to a complete breakdown of its security system within 60 to 90 minutes [1] - Attackers utilized approximately 17,000 zombie accounts to create live streaming rooms, broadcasting a large volume of illegal content, with some streams nearing 100,000 viewers [1] - The core reason for the extensive damage was identified as the transition of black and gray industries into an "automated attack" era, while Kuaishou relied on traditional manual defense methods [1] Group 2 - Experts emphasized that network security upgrades should not only focus on external attack defenses but also address risks from internal vulnerabilities [2] - Incidents of data leaks, internal account theft, and unauthorized operations have become frequent, with some attacks leveraging internal personnel and permission vulnerabilities [2] - Companies are advised to adopt a "defense against both internal and external threats" approach, integrating internal defenses into the overall security system, particularly focusing on "insider threat" prevention and permission management [2]
Google Wants More Energy to Fuel AI. It's Buying This Company to Help Power Its Data Centers.
Investopedia· 2025-12-22 20:05
Why This Is Significiant Like many of its Big Tech peers, Alphabet has said it plans to spend billions of dollars to raise its AI capacity. Monday's deal underscores how that could involve buying other companies, as well as more partnerships. Key Takeaways Google parent Alphabet just struck another big energy deal aimed at growing its AI data center footprint. Alphabet (GOOGL) said Monday it's buying energy infrastructure provider Intersect for $4.75 billion to help support its data center buildout. CEO Sun ...
Internet Giants ETF Captures AI Monetization Wave
Etftrends· 2025-12-22 18:47
Core Insights - The ALPS O'Shares Global Internet Giants ETF (OGIG) is experiencing a shift in artificial intelligence investment focus from hardware to revenue generation through advertising and data licensing [1] Performance Summary - OGIG has returned 29.1% over the past three years, outperforming the category average of 24.2% [2] - The fund has gained 15.6% year-to-date and 14.3% over the past year [6] Company Highlights - AppLovin Corp. (APP), which constitutes 2.4% of OGIG, saw a significant surge of 105.25% in Q3 after expanding its AI advertising engine, Axon, into e-commerce and other sectors [3][4] - Reddit Inc. (RDDT), holding 1.7% of the fund, climbed 52.75% in Q3 due to revenue growth from advertising and data licensing deals [5] Sector Contributions - Communication services provided the best sector performance for OGIG in Q3, contributing 3.8% to returns [7] - Information technology added 2.2% and consumer discretionary contributed 1.2% to the fund's performance [7] Fund Composition - The fund holds $141.8 million in assets under management with an expense ratio of 0.48% [6] - Meta Platforms Inc. (META) is the largest holding at 6.2%, followed by Microsoft Corp. (MSFT) at 6.1% and Alphabet Inc. (GOOGL) at nearly 6% [6]
The AI power scramble intensifies with Alphabet's purchase of data-center partner Intersect
MarketWatch· 2025-12-22 18:35
With energy becoming a critical AI bottleneck, Google's $4.75 billion acquisition of Intersect is a move to secure power for its data centers ...
Alphabet is buying Intersect: is it a buy signal for GOOGL shares?
Invezz· 2025-12-22 17:46
Alphabet Inc (NASDAQ: GOOGL) is in focus today after announcing plans to acquire Intersect, a leading clean energy and data centre infrastructure firm based in San Francisco, CA. Integrating Intersect's clean power generation capabilities with its cloud infrastructure will enable Google to scale more efficiently while reducing exposure to volatile energy markets. The acquisition also strengthens its ESG profile – appealing to institutional investors increasingly focused on sustainability. Beyond optics, the ...