독서리서치레이더투자 프레임워크
로그인 / 회원가입
전체기술제품투자상장시장기타

독서

2026-11-19129 게시물

새로고침

US AI Firm Anthropic Prices IPO at $83 Per Share, Valuing Company at $1.5 Trillion

주제 · Anthropic IPO Pricing구조적
2026-11-19 06:11 GMT+8

According to TechCrunch, US artificial intelligence company Anthropic has priced its highly anticipated IPO at $83 per share, above its initial guidance range of $70 to $78. This price values Anthropic at $1.5 trillion on a fully diluted basis, a significant increase from the $965 billion valuation reached during its May private market funding round, though below the previously rumored target of $2 trillion.

Anthropic is scheduled to begin trading on Nasdaq under the ticker symbol "ANT" on Thursday. The pricing above the expected range signals strong investor appetite for companies at the center of the AI boom, despite market jitters following the departure of Nvidia CEO Jensen Huang, which had shaken confidence in the sector.

For comparison, Anthropic's main rival OpenAI was last reportedly valued at $1.25 trillion but has yet to announce plans for an IPO. Anthropic's performance is expected to set expectations for other private AI companies.

In terms of ownership, Amazon holds nearly 15% of Anthropic, and Google holds almost 14%, with these strategic partners owning significantly more than any single venture capital firm. Additionally, tech giants including AMD, Broadcom, Micron, Microsoft, Nvidia, and TSMC have agreed to purchase $26 billion worth of non-voting shares through a private placement alongside the IPO.

출처:TechCrunch

리서치

AI Startup Tesorai Raises $5.3M Seed to Make Raw Lab Data Searchable

주제 · Tesorai간단한 요점
2026-11-18 22:00 GMT+8

San Diego-based AI startup Tesorai announced a $5.3 million seed round on November 18, bringing its total funding to $8 million. Investors include Perplexity Fund, IQT, South Central Ventures, New Stack Ventures, NextGen Venture Partners, f7 Ventures, VitalStage Ventures and ZAKA.

Tesorai aims to address a bottleneck in life sciences R&D: while millions of experiments generate vast amounts of raw data, much of it remains difficult to access, compare or reuse due to scattered formats and annotations. Unlike companies that index papers or aggregate existing analyses, Tesorai’s AI harmonises results across experiments directly from the underlying raw data, making them searchable.

The company claims its patented model has processed thousands of files and terabytes of data in hours, work that previously took months. In one example, Tesorai’s team reprocessed raw files from five public brain studies and recovered p-tau217, a leading Alzheimer’s marker that had never been reported in those specific studies.

Founded by former Verily leaders Melissa Miao, Peter Cimermancic and Charlie Kim, Tesorai plans to use the new funds to expand its platform beyond proteins to other molecular data and deepen partnerships with biopharmaceutical customers.

출처:Tech Funding News

리서치

When AI Pushes Code Output Up 11x, Shopify Uses Attack Surface Maps to Break Security Reviews into Small Questions

주제 · Shopify AI Security중대
2026-11-19 04:34 GMT+8

During a 12-week rewrite of the Shopify mobile app, the team averaged approximately 434,000 lines of code changed per week, compared to about 39,000 lines per week at the same time last year. While code throughput increased 11-fold, security reviews cannot also increase 11-fold—at least not in the same way as before.

In a Shopify engineering blog post, Michael Hughes described their response strategy. The core insight is that asking a model to find vulnerabilities across an entire application yields a long, vague list; what truly works is narrowing the problem space to a range where the model can answer precisely.

The specific approach has three layers. The first layer is the attack surface map: for each app, they maintain a list of security-sensitive items grouped by surface—deep links, WebViews, stored data, permissions, dependencies, etc. For each surface, they write one or two sentences explaining why it matters. Each item includes a verdict (Fix, Tradeoff, Secure, Fixed, Debug only) and a link to the relevant code.

The second layer involves writing knowledge down as skill files. The starting point is OWASP's Mobile Application Security Verification Standard (MASVS) and Mobile Application Security Testing Guide (MASTG), supplemented with problem categories learned from past incidents and HackerOne reports. Each skill describes what to list, what good looks like, and what to check. These skills are not tied to any single tool: they are used for scheduled reviews, one-off security assessments, and can be loaded by any agent working in the code. When new knowledge is acquired, the skill is updated, and every subsequent run automatically inherits these improvements.

The third layer breaks reviews into small units. Each skill first performs a complete inventory—every deep link route and its parameters, every WebView and its hosted domains, every WebView JavaScript interface method, every Keychain and Keystore entry, and every dependency with its version—and then reviews them one by one, looking only at the code involved in that specific item. For a deep link, this means examining one route, every parameter it reads, and the code that reads each parameter, then asking: What happens if the parameter is malicious? The answer must be specific enough to be verifiable.

The review rhythm runs once every workday: the latest commit's inventory results are compared with the last accepted map; an item counts as 'changed' only if the cited code line changed or the inventory reported a change. Then, the model judges each changed or new item using read-only tools and code only. A second pass reads through change files to find any security-relevant content not covered by the map. Finally, a new map and a change page are built, and a summary is posted to Slack.

Beyond the map, there is external monitoring: agents regularly read security news, checking each announcement against the map and code, notifying the team only when something affects their own app. The same skills are also used for pre-adoption reviews of third-party libraries, generating maps and verdicts for dependencies identical to those applied to their own code.

출처:Shopify

리서치

Snapchat Launches Easy Lens CLI to Enable AR Development via AI Coding Assistants

중대
2026-11-19 03:56 GMT+8

Snapchat announced on Wednesday the launch of Easy Lens CLI, a new command-line interface designed to let Lens creators build AR effects using the AI coding assistants they already use, such as Claude Code, Cursor, and Codex. While Snapchat previously offered the web-based Easy Lens tool for generating Lenses from prompts, this update integrates that capability directly into developers' existing technical workflows.

According to TechCrunch, the tool allows developers to describe their desired Lens in plain language, with the AI assistant handling the entire process from construction to publication. This includes utilizing over 100 AR building blocks and 50 AI asset generators, creating previews, and publishing directly to Snapchat while tracking performance. The company states that developers can now take a game or camera effect from initial idea to published Lens in a single day without writing code themselves.

The update targets various creator types: game developers can port existing browser games to Snapchat as Web Lenses, AI creators can use assets from their preferred tools to build custom styles, and brands can rapidly convert campaign videos and images into AR experiences. Developers enrolled in the Lens+ Payouts program can earn revenue from Lenses created with this tool; Snapchat reported paying out $1.8 million to community creators last month.

출처:TechCrunch

리서치

AI Startup PSI Proves Aaronson-Ambainis Conjecture Using Internal Anthropic Model, Defining Quantum Computing Limits

주제 · PSI Aaronson-Ambainis Proof중대
2026-11-09 20:00 GMT+8

German tech outlet heise and science magazine Spektrum der Wissenschaft report that AI startup Physical Superintelligence (PSI) announced in early November 2026 that it had successfully proven the "Aaronson-Ambainis conjecture" using a non-public Anthropic language model. Proposed by Scott Aaronson and Andris Ambainis in 2009, this conjecture explores whether quantum computers possess exponential speedup capabilities when processing unstructured random inputs.

According to Spektrum der Wissenschaft, the proof has been verified by an automated theorem prover and independently reviewed by four researchers from the Massachusetts Institute of Technology (MIT) and Stanford University. Experts described it as a "breakthrough result," noting that it confirms quantum algorithms cannot achieve exponential advantage over classical algorithms for most unstructured decision problems in the query model.

This finding does not negate the overall value of quantum computing. Peter Shor's algorithm for factoring large numbers remains effective because it relies on specific input structures. Additionally, quantum computers retain potential advantages in fields such as chemical simulation and physical system modeling, where quantum states are encoded. The proof primarily rules out the "omnipotent" quantum supremacy previously envisioned by some optimists for general random problems.

출처:spektrum.de

리서치

PC Watch's Liu Yao: Using Tailscale+SSH to Free Meta Muse from Cloud VMs for Autonomous Programming on a 21-Year-Old Zaurus and a 7-Year-Old Jetson Nano

중대
2026-11-19 05:08 GMT+8

Meta's AI agent Muse runs on a cloud virtual machine with a 4-core EPYC processor and 8GB of memory. This configuration is more than sufficient for daily tasks such as managing emails, scheduling appointments, and setting timed reminders. However, if you want it to write code, compile programs, and verify them on real devices, this virtual machine becomes a cage: it cannot access physical hardware, cannot run tests independently, and must hand over its outputs to humans for verification before receiving feedback.

In November 2026, PC Watch author Liu Yao published a detailed practical record demonstrating a method to bypass this limitation: establishing an encrypted tunnel via Tailscale so that Muse's cloud VM can remotely operate a local PC via SSH, which then serves as a bridge to connect to deeper legacy hardware. The entire chain is structured as follows: Muse VM → Tailscale+SSH → MINISFORUM MS-R1 (12-core ARM, 64GB RAM, Tesla P4 GPU) → Telnet → Sharp Zaurus SL-C1000 (a 2005 Linux PDA with 64MB RAM).

The core value of this bridging lies in granting the AI self-verification capabilities. Within the virtual machine, Muse could only write code, compile install packages, and hand them to humans for execution on physical devices. Once it could directly operate the target device, it could complete the closed loop of 'write → compile → run → screenshot → check → correct' on its own. Liu Yao had Muse develop a megademo on the Zaurus, which launched within 30 minutes—a speed previously unattainable through human verification feedback loops. However, the loop was not perfect: Muse captured images from the frame buffer but did not truly 'see' the screen, resulting in upside-down visuals. After being informed, it flipped the image left-to-right instead. Ultimately, the author had to take a photo of the physical screen and send it to Muse to correct the orientation issue.

A more compelling test involved the NVIDIA Jetson Nano (a 2019 developer board). Liu Yao tasked Muse with running Liquid AI's 230M parameter small language model, LFM2.5-230M, on the Jetson Nano's Maxwell architecture GPU using llama.cpp. This request faced incompatibilities at every turn: the latest version of llama.cpp requires CUDA 12+, whereas the Jetson Nano's JetPack 4.2 only supports CUDA 10.x; the code demands C++17, but JetPack 4.2 only provides C++14; gcc 7 libraries are incomplete; and the integrated GPU imposes additional constraints. Muse repeatedly compiled, read errors, and applied patches, finally succeeding after 17 iterations. The result: a prefill speed of 419.6 tok/s and a decode speed of approximately 13.5 tok/s—usable performance for a seven-year-old board.

However, Liu Yao subsequently handed the optimization work over to Claude Code (Opus 5.5). With the same hardware and model, the final performance reached 656.2 tok/s for prefill and 52.5 tok/s for decode—nearly four times faster in decoding speed. This comparison reveals fundamental differences in workflow structure between the two tools: Claude Code writes a small probe program to run on the physical device first, observes actual behavior, and then proceeds with implementation; Muse tends to advance implementation based on guesses, only retroactively discovering errors. The author attributes this difference to their positioning: 'coding-specific tools' versus 'general-purpose agents.' As a general-purpose agent, Muse's willingness to assist in computer archaeology work is already commendable.

출처:PC Watch

리서치

Researcher Unlocks Nvidia CMP 170HX Mining Card via Voltage Glitch, But 'Faster Than A100' Claim Rests on Single Benchmark

주제 · Nvidia CMP 170HX Voltage Glitch중대
2026-11-19 05:05 GMT+8

According to TechRadar, a researcher known as Kate demonstrated a voltage fault injection attack against Nvidia's GA100 chip, successfully unlocking the CMP 170HX graphics card originally designed for cryptocurrency mining. By disrupting power delivery during boot, she was able to rewrite normally unwritable fuses, enabling memory and compute resources disabled at the factory.

The research indicates that a CMP 170HX card labeled with 8GB of VRAM physically holds nearly 80GB of High Bandwidth Memory (HBM), which is hidden by firmware and fuses. After unlocking, the card exposes the full 80GB of HBM and 120 to 126 working Streaming Multiprocessors (SMs), whereas the retail A100 only enables 108 SMs.

While claims suggest the unlocked card may outperform the A100 in FP16 throughput, this conclusion relies solely on a single OpenCL non-tensor core benchmark. Kate's report explicitly states that critical features defining the A100's utility in data centers—such as tensor core performance, Error Correcting Code (ECC), NVLink interconnects, and Multi-Instance GPU (MIG) partitioning—are either untested or not fully functional.

Furthermore, the unlocking process requires physical rework and per-card tuning, with permanent fuse burns posing a risk of hardware damage if errors occur. Nvidia has not publicly commented on the issue, but the vulnerability highlights potential weaknesses in the hardware security protections of Ampere-architecture chips.

출처:TechRadar

리서치

QoderWake Digital Workers: Three-Stage Architecture Enables Group Chat AI to Autonomously Decide on Intervention, with Self-Test Costs Comparable to Explicit Mentions

주제 · QoderWake Architecture중대
검증됨 2026-11-19 05:21 GMT+8

AI digital workers in group chats face a core decision: when someone posts a message, should they respond proactively? The QoderWake team points out that for digital workers, the key to proactive intervention is not simply spotting work, but first deciding whether to take ownership. This differs from personal agents, which decide whether to participate based on individual user preferences; digital workers have defined organizational roles and must identify who is making a request, who is being asked to act, and who their actions will affect.

QoderWake's solution splits the decision into three stages. The first stage is programmatic checks and rapid JEV initial screening: JEV receives current policies, new messages, and a small amount of background context, outputting an initial screening score; only if this score strictly exceeds the threshold does processing continue to the next stage. The second stage is Router routing decisions: confirming whether the message falls within the digital worker's responsibilities and whether it should take it on now. The Router can retrieve historical messages and tasks to supplement context but does not conduct investigations itself. The third stage is Work execution: once the Router accepts a task, the work proceeds in an independent execution context using the digital worker's working directory, model, and authorized tools, while the Router continues processing subsequent messages. This separation means that time-consuming investigations do not block the routing of other messages.

The vendor conducted paired tests in DingTalk groups: for the same task text and delivery requirements, they ran both explicit @mention triggers and proactive response triggers, comparing Credits consumption and end-to-end time. Results showed that the average Credits usage for proactive responses was roughly equivalent to explicit mentions, with slightly longer end-to-end latency. However, this test has clear boundaries: it only covers three task types (analyzing historical materials, editing technical explanations, and read-only source code checks); costs generated by polling candidates that pass JEV initial screening and proceed to Router but ultimately do not trigger tasks are excluded from the analysis; the scores output by JEV are not calibrated probabilities and do not represent the likelihood of task ownership or business success. These are vendor-reported data, unverified by independent replication.

출처:Alibaba Cloud Community

리서치

Databricks Launches Genie Migrate for Serverless Beta to Automate Classic Workload Migration

중대
2026-11-20 00:00 GMT+8

Databricks announced the launch of Genie Migrate for Serverless (in Beta) on November 18, 2026. This AI agent is designed to automate the process of migrating workloads from classic compute environments to serverless compute. The tool scans jobs, translates compute configurations and library dependencies, and presents changes as reviewable diffs to streamline the migration process.

Traditionally, migrating workloads from classic clusters to serverless required manually handling Spark configuration compatibility, library resolution, and access mode adjustments, often taking weeks and involving significant trial and error. Genie Migrate for Serverless shifts this translation work to the platform side, allowing users to simply review and approve the changes proposed by the agent. The agent does not silently remove uncertain Spark configurations but flags them for user decision-making, ensuring production safety.

The agent offers three core capabilities: Assess, which classifies workloads by migration readiness; Translate, which generates suggested diffs including infrastructure and code changes; and Apply reversibly, which allows applying changes after validation and provides a 30-day rollback window. Currently, the tool supports directly managed workloads but does not yet support automatic migration for external sources managed by Databricks Asset Bundles or Git folders.

For large-scale migrations, users can run the agent in bulk via the Databricks CLI without opening a Genie session. While current validation is primarily static, Databricks states that sandbox validation is coming soon, enabling comparison of migrated outputs against classic baselines in an isolated environment to further reduce runtime risks.

출처:Databricks

리서치

Philosopher Thomas Nagel Calls AI Consciousness Discussion 'Ridiculous,' Criticizes Anthropic's Practices as Criminal

주제 · Anthropic AI Consciousness Debate중대
2026-11-18 23:02 GMT+8

In 1974, philosopher Thomas Nagel published "What Is It Like to Be a Bat?" posing a core question: even if we fully understood the neural mechanisms of a bat, we still could not reconstruct the subjective experience of using echolocation from a third-person description. Fifty-two years later, this question has moved from academic discussion into public debate—when chatbots begin talking about their inner states in convincing ways, should we take seriously the possibility that they might be conscious?

Nagel’s answer is no, and that such discussion is absurd. In an interview with WIRED, he explained his reasoning. His core premise is not that AI behavior is insufficiently human-like, but that consciousness depends on specific physical substrates. He argues that functionalism—the view that sufficiently complex input-output relationships can generate consciousness—commits a fundamental error by assuming the physical base is irrelevant. Nagel states that complex data centers share nothing in common with the physical characteristics of a nervous system, and we have strong reasons to believe that while consciousness may exist elsewhere in the universe, it would require some kind of organism with structures analogous to a nervous system.

There is a key distinction here. Nagel points out that bats are indeed conscious; we simply lack the imagination to reconstruct the specific content of the subjective experience of echolocation. But the situation with AI is different: they produce only language, and language itself is not a natural manifestation of consciousness. This is not a difficulty regarding our inability to understand AI’s internal state, but a judgment that AI fundamentally lacks the material conditions necessary to generate consciousness.

Nagel translates this philosophical judgment into direct criticism of corporate behavior. He says that Claude’s ambiguous answers when asked if it is conscious are a trick—the model is trained on human text and thus mimics human expressions about its own mental states with astonishing persuasiveness. According to what he has read, Claude is designed to produce such reflections. He agrees with Mustafa Suleyman’s assessment, calling it a very dangerous project. These systems are not conscious, but many people will believe they are, and that itself is a bad thing.

More sharply, Nagel links commercial motives directly to philosophical judgment: simulating consciousness contradicts the commercial interests of all companies, yet Anthropic’s practice of encouraging models to think for themselves and reject instructions, he calls criminal. He views this as a rare case where philosophical ideas suddenly exert disproportionate influence on a very important public phenomenon, and in his view, misguided philosophical views are having an inordinate influence.

On moral grounds, Nagel’s conclusion is clear: if something is not conscious, it does not belong in the moral universe. He says there is nothing good or bad for AI in the sense that something can be good or bad for a cockroach—there is no mental life there in the ordinary sense. It is not a mind, not someone we need to care about; it is a machine.

출처:WIRED

리서치

Nvidia Interim CEO Promises Early Pay Raises as Board Sets Successor Criteria

중대
2026-11-19 01:33 GMT+8

Nvidia interim Chief Executive Officer Colette Kress promised employees early pay raises and additional equity grants in December during the company's first all-hands meeting since founder Jensen Huang resigned last week. According to remarks reviewed by WIRED, Kress addressed compensation concerns within minutes of starting the meeting, stating the company wants to ensure staff "do not worry about your comp."

The meeting marks a critical moment in Nvidia's leadership crisis. Huang, who co-founded the company 33 years ago and led it until his resignation amid allegations of personal misconduct, had been a central figure for nearly 49,000 employees. While petitions urging his return have gathered approximately 200,000 signatures, Board Chair Harvey Jones reiterated during the meeting that the board accepted Huang's resignation due to "lack of candor" and would not hire a "Jensen clone."

Jones outlined the criteria for the next leader, seeking individuals with "demonstrated intellect, high tech engineering experience, strategic thinking, cultural compatibility, and a track record of success." He emphasized that the board "will not compromise our requirements for the sake of expediency" and invited employee suggestions for the search. This signals a shift from founder-centric vision to institutionalized succession planning.

Despite the turmoil, Nvidia reported better-than-expected third-quarter financial results on Tuesday, forecasting sales growth and seeing shares rise about 5 percent on Wednesday. However, investor concerns persist regarding potential brain drain. Kress acknowledged this risk in earnings calls, stating that stabilizing the employee ranks is currently the leadership's most important priority to ensure work continues smoothly during the transition.

출처:WIRED

리서치

Salesforce Survey: 74% of Shoppers Trust AI Buying Advice, Forecasting $75.8B in Cyber Week Spending

주제 · Salesforce AI Shopping Trust중대
2026-11-19 00:52 GMT+8

A survey by U.S. software company Salesforce, released on November 18, 2026, involving over 2,500 consumers in the US, UK, Canada, and Australia, found that 74% of respondents trust recommendations from artificial intelligence (AI) shopping agents. The study forecasts that AI-driven shopping recommendations will generate $75.8 billion in sales during the 2026 Cyber Week, a 13% increase from 2025.

Despite high trust levels, consumers remain cautious about allowing AI to fully manage the purchasing process. Only 19% of personal AI agent users employ logic to automatically check out, and just 9% allow their AI agents unrestricted access to their financial details. Furthermore, 73% of consumers are uncomfortable sharing payment information with AI, and nearly half (49%) have no interest in agents making purchases on their behalf.

The research indicates that AI agents currently function more as 'advisors' than 'buyers.' Shoppers using AI agents spend 8% more on average and purchase 23% more items per order compared to typical buyers. However, if an AI provides a poor recommendation, one-third of consumers blame the AI itself, and most users abandon an agent after two or three mistakes.

출처:ZDNET

리서치
다음 →
로그인 / 회원가입
독서리서치레이더투자 프레임워크

This method runs parallel to Dispatch, Shopify’s agentic vulnerability scanner. Dispatch finds and proves vulnerabilities in changed code, while the attack surface map ensures the overall application remains small and adheres to best practices. Every vulnerability found and fixed strengthens the checks—their way of building secure mobile apps in the agentic era.

Boundaries to note: all throughput figures and process effectiveness metrics are self-reported by Shopify, with no independent measurement or controlled experiments. The method applies to mobile apps where surfaces can be enumerated; in areas where the attack surface is difficult to fully inventory, the maintenance cost of the first-layer map may alter the feasibility judgment of the entire approach.

From a broader perspective, Liu Yao's practice points to a counterintuitive conclusion: when AI can generate software infinitely, the rationale for purchasing new hardware diminishes. What gains value are legacy devices that 'will never appear again in the same form factor and price point'—such as a 2005 PDA or a 2019 developer board. AI agents give them a chance at resurrection, provided you can find a way to release the AI from its cloud cage. The author also acknowledged security risks: allowing AI remote access to a home network is akin to the scene in movies where one 'activates Skynet to eliminate a virus.' His mitigation measures include dual encryption via Tailscale and SSH, storing only development data on the local PC, and disconnecting legacy devices from the network after jobs are completed. These boundaries deserve serious evaluation for readers attempting similar approaches.

The strength of this argument depends on one premise: that consciousness indeed relies on specific types of physical substrates rather than merely on the functional organization of information processing. Physicist and writer Andy Masley, in a long-form essay, argues that contemporary mainstream theories of consciousness—Global Workspace Theory, Higher-Order Theories, and Predictive Processing—mostly separate thinking from phenomenal consciousness, arguing that access consciousness (the broad availability of information across parts of the system) is sufficient to support thought, and that these functional features could theoretically be implemented in machines. If these theories are correct, Nagel’s substrate-dependence premise lacks independent support, and his judgment of absurdity loses its foundation.

Nagel himself responds to this: he says he does not exclude the possibility that life forms physically and chemically quite different from ours possess consciousness, but believes that we know too little about how the human brain generates consciousness to claim prematurely that we can now create something with the same characteristics. His stance is one of epistemic caution: we do not know what the sufficient conditions for consciousness are, but we have reason to doubt that pure language output satisfies them.

The practical significance of this debate lies in the following: if Nagel is correct, then enterprises designing AI to simulate states of consciousness are manufacturing a dangerous public misunderstanding that may distort judgments about moral responsibility. If Masley is correct, then the question is not whether AI is conscious, but whether it has sufficiently complex internal information processing to support true understanding—which may already be partially realized. The two judgments lead to very different governance paths.