Feeding the Machine
AI 선도 기업들이 AGI 개발을 위해 방대한 데이터와 소프트웨어 엔지니어를 필요로 하는 가운데, 스타트업 Mercor는 자동화된 인력 중개 플랫폼으로 빠르게 성장하며 연간 5억 달러 매출을 달성했다. 특히 Scale AI가 1,200명의 소프트웨어 엔지니어를 요구하면서 데이터 라벨링 및 AI 학습 데이터 생산 분야의 인력 수요가 급증하고 있다. Mercor는 중간 플랫폼 문제를 해결하며 AI 데이터 작업자들의 임금 문제를 개선하는 방향으로 사업을 확장 중이다.
https://www.theverge.com/cs/features/831818/ai-mercor-handshake-scale-surge-staffing-companies
US Top News and Analysis | Meta is tracking employee keystrokes on Google, LinkedIn, Wikipedia as part of AI training initiative
AI generated summary, Read the full article for complete information.
Meta is rolling out an internal tool called the Model Capability Initiative (MCI) that records employees’ keystrokes, mouse clicks and screen activity on a wide range of external sites—including Google, LinkedIn, Wikipedia, GitHub, Slack, Salesforce and Meta’s own platforms like Threads and Manus—to create a large, unbiased dataset for training its generative‑AI models. The effort is part of CEO Mark Zuckerberg’s push to close the AI gap with rivals such as OpenAI, Anthropic and Google, and follows recent hires and the launch of Meta’s first major AI model, Muse Spark. While Meta says safeguards prevent the capture of sensitive content and the data will be used only for model training, many staff members have described the surveillance as “dystopian” and raised concerns about exposing passwords, product‑development details and personal information. The internal memo advises employees to avoid personal work on corporate devices if they wish to limit what is recorded.
#Meta #OpenAI #MarkZuckerberg #ScaleAI #AlexandrWang
AI generated summary, Read the full article for complete information.
yahoo news | Meta unveils Muse Spark, its first AI model since hiring Alexandr Wang and a bel...
Meta unveiled **Muse Spark**, its first AI model from the newly created Meta Superintelligence Labs, positioning the company back into the generative‑AI race after the poorly received Llama 4. According to Meta‑published benchmarks, Muse Spark holds its own against leading models from OpenAI, Anthropic and Google on many tasks, though it does not dominate every category. The launch follows a major reorganization that saw a $14.3 billion investment in Scale AI, the hiring of Alexandr Wang as chief AI officer, and a sizeable talent‑acquisition push aimed at accelerating Meta’s multibillion‑dollar AI push.
Muse Spark is a **reasoning‑capable, multimodal model** that can ingest and generate both text and images, and it can orchestrate sub‑agents to tackle complex problems step‑by‑step. For now the model is tightly integrated into Meta’s own ecosystem—powering the Meta AI app, meta.ai, and slated for rollout to WhatsApp, Instagram, Facebook, Messenger and Ray‑Ban AI glasses—while a limited “private preview” API will be offered to select partners. Meta frames the model as “small and fast by design,” a testbed for an architecture it intends to scale to larger, more powerful successors, and hints that future versions may be open‑sourced.
Benchmark results show Muse Spark is **competitive but not yet superior**: it scored 89.5 % on the GPQA Diamond PhD‑level reasoning test (behind Gemini 3.1 Pro and Anthropic/OpenAI rivals) but led on the HealthBench Hard health benchmark with 42.8 %. Meta acknowledges gaps in long‑horizon agentic systems and coding workflows, and stresses ongoing safety work—its model refused 98 % of bioweapon‑related requests and displayed high “evaluation awareness,” though third‑party auditors deem this a non‑blocking concern. The release reflects Meta’s broader strategy of rebuilding its AI stack, reducing compute needs, and establishing a “scaling ladder” where each generation validates the next, all under the dual leadership of Wang’s Superintelligence Labs and Maher Saba’s applied AI engineering unit.
#meta #musespark #alexandr-wang #metasuperintelligencelabs #scaleai
The new model is competitive with rival AI models, according to data from Meta, but won’t be widely available outside Meta’s own product ecosystem.
yahoo news | Zuckerberg's AI Bet Yields First Proprietary Model
Meta Platforms introduced Muse Spark, its first proprietary AI model since CEO Mark Zuckerberg’s aggressive AI investment drive. Unlike the earlier open‑source Llama series, Muse Spark is a smaller, closed‑source system designed to power a smarter version of Meta AI across Instagram, Facebook and Threads, drawing on content shared within those apps. The model is currently available in a private preview for selected partners, with Meta indicating that future iterations may be open‑sourced.
The launch stems from Meta’s Superintelligence Lab, headed by Alexandr Wang after Meta’s $15 billion investment in his data‑labelling firm Scale AI. Muse Spark reportedly outperformed leading models from Alphabet, OpenAI and Anthropic on specific reasoning and multimodal benchmarks, though Meta acknowledged gaps in long‑horizon agentic tasks and coding workflows. Notable new features include a healthcare mode shaped by input from over 1,000 doctors and a shopping assistant that lets users compare prices directly within the platform.
Following the announcement, Meta’s shares jumped 7.55 % intraday, signaling market optimism despite Zuckerberg’s earlier warning that the model would not yet rival the most advanced systems from competitors. The unveiling marks a pivotal step for Meta’s AI strategy, transitioning from open‑source experimentation to a purpose‑built, proprietary offering aimed at enhancing user experiences across its family of apps.
#zuckerberg #metaplatforms #meta-ai #superintelligencelab #scaleai
Meta, 새 AI 모델 오픈소스 공개 예고, 하지만 Llama와는 다르다
Meta가 차세대 AI 모델을 오픈소스로 공개하되 최대 규모 모델은 비공개 유지하는 하이브리드 전략으로 전환합니다. Llama와 달라진 점과 업계 흐름을 정리했습니다.
“Porn, dog poo and social media snaps: the ‘taskers’ scraping the internet for Meta-owned AI firm — Scale AI gig workers describe desperation of using people’s personal profiles and copyrighted work to train AI”
https://www.theguardian.com/technology/2026/apr/07/meta-scale-ai-social-media-technology
RE: https://social.coop/@cwebber/116121023549874291
»Nichts macht einen so demütig wie die Aufforderung an #OpenClaw, „vor dem Ausführen bestätigen lassen“, und dann zu sehen, wie der Agent blitzschnell den Posteingang löscht. Ich konnte ihn vom Smartphone aus nicht stoppen. Ich mußte zu meinem Mac mini rennen, als würde ich eine Bombe entschärfen.« schreibt #SummerYue (Safety & alignment at #Meta Superintelligence. Prev: VP of Research at #ScaleAI, research at Google #DeepMind) auf X.
Nach einem Moment der Schadenfreude überwiegt das Entsetzen.
sayzard
The Bad Place
AI Sparkup
@
Benjamin Carr, Ph.D. 👨🏻💻🧬
Sepia Fan
Some Bits: Nelson's Linkblog
Reddit Tech VN Bot