LecturaInvestigaciónRadarMarco de inversión
Iniciar sesión / Registrarse
TodoTecnologíaProductosInversiónBolsaOtros

Lectura

2026-11-0499 publicaciones

Actualizar

Prompt Engineer Susanne Renate Schneider: AI Skills Should Be Manually Triggered First, Automate After Verifying Timing

Tema · CustomGPTsResumen rápido
2026-11-04 20:30 GMT+8

Skills in ChatGPT and Claude are not more complex prompts; they are reusable work instructions. A Skill specifies the concrete steps for a certain type of task, the expected output, and the context in which it should be invoked. Creating one requires no programming skills; natural language descriptions suffice. Prompt engineer Susanne Renate Schneider believes the standard for judging whether a Skill is worth the effort is straightforward: Are you repeatedly performing the same type of task, always relying on the same criteria for judgment?

While Skills can be set to trigger automatically, Schneider recommends starting with manual triggers. The reason is: only after you have personally invoked them a few times can you see which scenarios were meant for the Skill to intervene automatically and which merely occur by chance. She suggests configuring Skills to require explicit commands to start initially, observing for a period, and then gradually enabling automation, adjusting the trigger conditions at that point.

For users unsure where to begin, Schneider shares a starter prompt she has used herself: 'Of the tasks I complete here in [insert specific AI tool], which are worth creating a Skill for?' This prompt relies on the AI tool's memory function being enabled; its role is to filter through your existing conversation history to identify recurring task patterns, helping you find the first batch of scenarios worth solidifying into Skills.

Notably, CustomGPTs in ChatGPT previously allowed for similar workflow encapsulation, but OpenAI has announced it will cease this service on December 11. Users can migrate existing CustomGPTs to plugins; specific instructions are provided in the November 5 issue of the t3n MeisterPrompter Newsletter.

Fuentes:t3n.de

Investigación

Freshworks Crosses $1 Billion ARR Ahead of Schedule as Q3 Turns Profitable

Tema · FreshworksMaterial
2026-11-04 20:46 GMT+8

US software company Freshworks announced in its third-quarter fiscal 2026 report that its Annual Recurring Revenue (ARR) has crossed the $1 billion threshold, achieving this milestone ahead of the end-of-year target set by CEO Dennis Woodside. According to diginomica, the company reported a GAAP net profit of $18 million for the quarter ended September 2026, compared to a loss of $4.6 million in the same period last year. Total revenue reached $245 million, representing a 14% year-on-year increase.

This financial milestone marks substantial progress in Freshworks' strategic transition from a traditional customer support tool provider to an enterprise IT Service Management (ITSM) and Employee Experience (EX) platform. Data shows that ARR in the EX vertical grew by 22% on a constant currency basis to $588 million, now accounting for 59% of the company's total ARR. In contrast, Customer Experience (CX) ARR stood at $404 million, growing only 4% on a constant currency basis.

CEO Dennis Woodside attributed these results to the company's strategy of "moving upmarket," focusing on large enterprises and high-value customers rather than broadly targeting small and medium businesses as in the past. He noted that customers choose Freshservice for its unified data model and platform capabilities across IT, HR, and finance, allowing it to replace legacy systems. Additionally, the company acquired Torii and Robusta last month to enhance its AI capabilities in IT asset management and incident investigation, further solidifying its position in enterprise operational data governance.

Fuentes:diginomica

Investigación

Apple's Low-Capital AI Strategy: Not Building All Models, Investing Less in Infrastructure, Profiting from Interface Position

Tema · 苹果AI战略Material
2026-11-03 23:24 GMT+8

During Apple's fiscal year 2026 fourth-quarter earnings call, AI and Apple Intelligence were mentioned 45 times, setting a historical record. But what truly matters is not the frequency of mentions, but the fact that the company demonstrated an AI economic model distinct from its peers: it does not need to train all large models itself, nor spend tens of billions of dollars building data centers; instead, it profits from AI by selling devices, collecting platform fees, and implementing planned subscription tiers.

The comparative figures are striking. Apple's capital expenditure for the last quarter was $2.2 billion, while Amazon reported $77.3 billion during the same period, and Alphabet (Google) reported $51.4 billion. These are company-wide capital expenditure figures, not limited to AI, and they exclude some AI-related expenses, yet the magnitude of the gap remains significant. Apple Chief Financial Officer Kevan Parekh acknowledged on the call that Siri AI's cloud costs will be reflected in service costs and may expand as usage grows—he also stated that it is currently impossible to predict how many requests will require server processing rather than on-device computation. This means Apple's low capital expenditure is not a permanent commitment, but rather a phase-specific cost structure choice.

Apple has three revenue pathways. The first is hardware: iPhone revenue grew 17.5% year-over-year to $57.6 billion, setting a September-quarter record, while Mac revenue grew 24% to $10.8 billion, its highest level in four years. Parekh described AI as "a compelling reason for customers to upgrade and switch to get the latest features." The second is platform commissions: according to data cited by The Wall Street Journal from analytics firm AppMagic, Apple is expected to collect over $1 billion in App Store fees from generative AI apps in 2026. The third is upcoming usage-based subscriptions: newly appointed CEO John Ternus stated that features relying on server-side models (including Siri AI and intelligent photo editing) will have usage limits, and customers needing more capacity "can purchase expanded access in the future." Ternus illustrated this logic's evolution using iCloud as an example—"Initially, iCloud tiers were primarily about storage, but we have been adding more capabilities and features."

Mac's local AI capabilities represent a specific pillar within this strategy. During the earnings call, Apple highlighted two enterprise cases: Gap runs an AI loss-prevention system entirely on Mac Mini computers across hundreds of stores, avoiding cloud token fees; Christie's auction house uses local models on Mac Studio computers to optimize workflows such as pricing estimates. Ternus said the Mac Mini and Mac Studio "are pushing the boundaries of local private AI, enabling users to do things once considered impossible." Former CEO Tim Cook previously noted that customer recognition of the Mac Mini and Mac Studio's capabilities in AI and agentic tools had exceeded Apple's expectations.

However, this model has a clear vulnerability. Apple's platform commission revenue depends on customers paying through its systems. According to The Wall Street Journal, OpenAI has begun directing potential subscribers to pay directly through its website, bypassing App Store commissions. Apple's position allows it to profit from competitors' AI, but those competitors similarly have incentives to circumvent its fee channels. Furthermore, Apple has not disclosed how much of its revenue growth reflects AI demand versus price increases, so the causal strength of AI-driven hardware upgrades remains an unquantified assumption.

Fuentes:The Verge

Investigación

Tesla and SpaceX Shares Drop Over 4% Premarket After Musk's Election Bet Backfires in Texas

Material
2026-11-04 22:41 GMT+8

Shares of U.S. electric vehicle maker Tesla and aerospace company SpaceX fell more than 4% in premarket trading on Wednesday. This reaction followed a string of Democratic victories in the Texas midterm elections, including James Talarico defeating Republican nominee Ken Paxton for the Senate seat and Gina Hinojosa unseating Governor Greg Abbott.

According to Business Insider, Elon Musk spent approximately $224 million backing Republican candidates in Texas, with $158 million of that total coming in September alone. With both companies headquartered in Texas, the Democratic win signals a potential return to regulation by Democratic officials. If the drops hold at market open, Musk's two public companies could lose more than $140 billion in combined market value.

Wedbush analyst Dan Ives noted that Musk "needs to find the path forward with Texas now with a Democratic win." Musk responded on social media, stating that SpaceX's plan to build data centers in space looked "pretty appealing right now," and jokingly added that his "election-interfering space lasers didn't work this time."

Fuentes:Business Insider

Investigación

Nigeria's Election Body Faces Pressure Over AI-Generated Data on Voter Roll

Tema · INECMaterial
2026-11-04 20:13 GMT+8

Nigeria's Independent National Electoral Commission (INEC) is facing intense pressure to address the discovery of AI-generated voter information and ID photos on a draft electoral roll. This finding has sparked serious concerns about the integrity of next January's presidential vote and risks further undermining faith in the politics of Africa's largest democracy.

According to Semafor, with just over 10 weeks until the vote, President Bola Tinubu is locked into a re-election battle against familiar rivals. More than 103 million Nigerians registered to take part in the poll, but INEC's numbers and overall preparations have come under intense scrutiny. Advocacy groups, technology analysts, and former INEC officials have led calls for an independent inquiry to identify how fabricated information came to be included. Many have warned that the false data would add to cynicism about the organizer's commitment to a fair election and weaken voter participation.

In response to the backlash, INEC said the contentious register is a temporary document open to review for errors and stated it would look into 'possible technical or procedural lapses' that allowed AI intrusions. A senior official blamed rogue staff, saying the commission 'doesn't know why' they would use their logins to tamper with the roll. However, analysts say these responses are unsatisfactory given the gravity of the situation. Kolawole Oluwadare, a senior official at Lagos-based rights group SERAP, told Semafor: 'This goes beyond what INEC can handle in house, which is why we are calling for an audit by an external body.'

Martin Uwakwe, a US-based technology analyst who reviewed the draft roll, found several errors in the register beyond the AI-generated issues, including duplicates and the presence of people who had died. He noted that the electoral commission could use Nigeria's existing identity database to authenticate voter data and ensure integrity. 'These things are very easy to do but I don't know why we are still having this conversation in 2026,' he told Semafor.

Resolving issues with the register is critical for voting to be credible on election day. With each election cycle, Nigeria has increased the role of digital technology, including biometric tests, in the voting process to improve transparency and efficiency. But the present qualms around AI manipulation of voter data may make people skeptical of the reported outcome. For example, the portable electronic devices used to accredit voters (known as BVAS) could be manipulated by those with backend access, potentially disenfranchising unwanted voters.

Fuentes:semafor.com

Investigación

Datadog Reports 35% Revenue Growth to $1.2 Billion in Q3, Launches First Specialized AI Model Scout 1.0

Material
Verificado 2026-11-04 20:29 GMT+8

Datadog, Inc. (NASDAQ: DDOG), a US-based observability and security platform, announced its financial results for the third quarter ended September 30, 2026, on November 4, 2026. The company reported revenue of $1.20 billion, representing a 35% increase year-over-year. GAAP operating income was $75 million, while non-GAAP operating income reached $334 million, corresponding to a 28% non-GAAP operating margin.

Cash flow metrics remained robust, with operating cash flow at $290 million and free cash flow at $251 million for the quarter. As of September 30, 2026, Datadog held $5.2 billion in cash, cash equivalents, and marketable securities. This performance highlights the company's ability to generate significant liquidity while maintaining high growth rates.

Customer expansion continued among larger enterprises. As of the end of Q3 2026, Datadog had approximately 4,860 customers with annual recurring revenue (ARR) of $100,000 or more, up 20% from about 4,060 such customers a year prior. This growth reflects sustained demand for unified monitoring solutions as companies scale their digital operations and AI deployments.

In addition to financial results, Datadog introduced Scout 1.0, its first specialized AI model developed by the Datadog AI Research team. Now powering Bits Investigation, Scout 1.0 is claimed to achieve frontier-model performance at a fraction of the inference cost, enabling more frequent and earlier investigations into production issues to reduce on-call burden and accelerate incident response.

For guidance, Datadog expects fourth-quarter 2026 revenue between $1.19 billion and $1.20 billion, with non-GAAP earnings per share between $0.69 and $0.71. Full-year 2026 revenue is projected to be between $4.515 billion and $4.525 billion.

Fuentes:sec.gov

Investigación

Airbus Deploys AI to Optimize Aviation Operations: Smart Catering Tracking and Predictive Maintenance Provide 10-Day Fault Warnings

Tema · 空客AI运营优化Material
2026-11-04 06:20 GMT+8

European aerospace giant Airbus is deeply integrating artificial intelligence and data analytics into core aviation operations, covering in-flight catering management, aircraft predictive maintenance, and ground technical support. This initiative aims to help airlines optimize meal planning, identify equipment anomalies earlier, and accelerate technicians' access to repair guidance through real-time data insights.

In catering management, Airbus partnered with Virgin Atlantic in 2025 to test the "Smart Catering" system. The system uses built-in cameras on existing crew tablets or smartphones to automatically recognize trays and update inventory records, reducing manual counting tasks. Citing International Air Transport Association (IATA) data, Airbus notes that the global aviation industry generates approximately 3.6 million tons of cabin waste annually, with 18% to 20% being uneaten food and beverages. Airbus states that the system's backend analysis has the potential to reduce avoidable in-flight catering waste by double-digit percentages.

In the maintenance sector, Airbus applies advanced analytical algorithms via its Skywise Fleet Performance+ platform to analyze technical data transmitted from flights to the ground. Airbus reports that the system can identify signs of anomalies up to 10 days before a failure occurs, allowing maintenance teams to schedule part replacements in advance according to engineering guidelines. This shift from reactive response to proactive prevention helps airlines reserve maintenance time and reduces the risk of subsequent flight delays caused by sudden breakdowns.

Additionally, Airbus's AIRTAC customer service center has introduced AI tools to assist engineers in retrieving vast amounts of technical documentation. According to Airbus, engineers can search thousands of pages in seconds to quickly locate troubleshooting procedures relevant to on-site issues, thereby shortening the wait time for ground maintenance personnel. Although actual repairs must still follow established procedures, improved information retrieval efficiency directly impacts the overall speed of troubleshooting.

Fuentes:TechOrange 科技報橘

Investigación

Crunchbase: Over Half of US VC Funding Flows to Seven Billionaire-Led Firms

Tema · 美国风投集中亿万富翁公司Estructural
2026-11-04 20:00 GMT+8

According to new data published by Crunchbase News, a significant structural shift occurred in the US venture capital market during the first ten months of 2026: over 50% of seed-to-growth-stage equity funding flowed to just seven companies. These firms are OpenAI, Anthropic, xAI (now part of SpaceX), Anduril Industries, Databricks, Prometheus, and The Boring Co. Together, they raised more than $260 billion in equity funding during this period.

Notably, the founders or CEOs of these seven companies are all billionaires. Elon Musk and Jeff Bezos, the world's wealthiest individuals, founded or lead multiple entities on this list. Musk is the founder of xAI and The Boring Co., and co-founder of OpenAI; Bezos launched the physical AI startup Prometheus in 2025, which raised $12 billion earlier this year. Other prominent figures like Sam Altman (OpenAI CEO) and Dario Amodei (Anthropic CEO) have also accumulated vast wealth through their equity stakes or prior investments.

This trend marks an evolution in the venture capital model. Historically, VC primarily funded first-time entrepreneurs or serial founders, with the term "unicorn" only entering common lexicon around 2013. Today, capital increasingly favors companies with valuations high enough to make stakeholders billionaires, or ventures founded by individuals who were already wealthy. This concentration implies that for most traditional startups, securing large-scale funding has become significantly harder, underscoring a growing Matthew effect in the capital markets.

Fuentes:Crunchbase News

Investigación

Canva Cuts 22% of Staff and Lowers Growth Forecast as AI Costs Drive Valuation Down to $35 Billion

Tema · Canva AI成本重组Material
2026-11-04 19:59 GMT+8

Australian design software company Canva has announced it is cutting 22% of its workforce, totaling 1,271 people, to address a slowdown in business growth experienced for the first time in a decade. Co-founder and CEO Melanie Perkins stated in a note to employees that while the company has achieved success over the past thirteen years, it must change to tackle slowing growth.

The restructuring includes merging product, growth, and platform teams to reduce communication friction, decreasing "subspecialisation," and requiring employees to work from the office a few days a week. Perkins described this as the company's "hardest year yet."

Canva has been deploying its own AI models to reduce reliance on third-party models and lower costs, but still had to cut its revenue growth forecast from 30% to 20% to address ballooning AI expenses. This adjustment led early investors to cut the company's valuation to approximately $35 billion earlier this year.

Additionally, Canva faces increasing competition from AI startups and general-purpose chatbots like ChatGPT and Gemini, whose image generation capabilities allow users to easily create posters and presentations.

Fuentes:TechCrunch

Investigación

ECB President Christine Lagarde: Short-Term GDP Impact of Climate Shocks Limited, but Financial Buffers Eroding

Tema · 拉加德气候金融稳定论Estructural
2026-11-04

Lagarde’s core judgment rests on a seemingly contradictory fact: ECB staff estimate that extreme weather in summer 2026 will have only a small and temporary effect on euro area economic activity, yet she believes the near-term picture understates the risk. This assessment is not based on reinterpreting short-term GDP figures, but on a more complete characterization of transmission mechanisms.

The first layer is persistent scar effects. Research shows that regions hit by heatwaves see economic activity average 1.5% lower two years later, while areas affected by droughts see activity average 3% lower four years later, partly due to declining productivity and population outmigration. This means that even if the immediate GDP loss from a single shock is small, cumulative effects alter the long-term trajectory of regional economies. Nearly three quarters of euro area businesses depend heavily on at least one service provided by nature, with water-related services being the most exposed.

The second layer is the insurance gap. Historically, only one quarter of catastrophe losses in Europe have been covered by insurance; the remaining costs are borne directly by households and businesses or ultimately passed on to fiscal authorities. When fiscal space itself is tight, this gap transforms from an economic issue into a sovereign credit issue. Nearly 12% of euro area economic output faces risks linked to insufficient flood protection.

The third layer is the vulnerability of energy buffers. At the beginning of 2026, wholesale electricity prices reacted less strongly to rising gas prices than during the 2021–2022 crisis, largely because euro area countries increased low-carbon power generation. Spain is a typical case: the share of hours in which gas plants set electricity prices fell from around 80% in 2020 to just above 20% in 2026. However, heatwaves and droughts simultaneously erode this buffer—low water levels reduce hydropower output, cooling water shortages limit nuclear capacity, and air conditioning demand drives up electricity load. The result is that the price cushioning provided by renewables and nuclear energy is weakened.

Notably, solar was the only major power source to perform better than usual during heatwaves, but evening supply tightened and prices peaked as solar output dropped while cooling demand remained high. This points to a specific investment gap: grids and storage. Lagarde cites data stating that every euro invested in upgrading the grid saves more than €2 in other system costs. Since new grid connections take two to ten years, she suggests allowing firms to adopt behind-the-meter self-consumption schemes, providing flexibility to the grid under strain in return.

For investors, Lagarde’s argument raises a question of time scale: the transmission of climate financial risks does not occur through a single GDP shock, but through the slow deterioration of regional economies, structural gaps in the insurance system, and the climate vulnerability inherent in energy transition achievements. Her three priorities—the EU Framework for Climate Resilience, grid and storage investment, and common rules allowing firm self-consumption—all point to a significant expansion in private sector financing needs. Lagarde explicitly states that most of these investments must come from the private sector, and completing the Savings and Investments Union while maintaining a stable climate policy framework, including carbon pricing, would lower financing costs.

Fuentes:European Central Bank

Investigación

Google Ties US School Search Access to AI Consent, Sparking Antitrust Criticism

Tema · 谷歌AI搜索绑定争议Material
Verificado 2026-11-04 18:49 GMT+8

Tech giant Google has adjusted its product strategy in the US education sector by linking access to its core service, "Google Search," with schools' consent to use artificial intelligence (AI) features. According to an announcement released on Tuesday, if school administrators choose to disable Google's AI tools, students will also lose the ability to perform searches through the Google ecosystem. This measure aims to boost the penetration rate of its AI technology in educational platforms but has sparked intense controversy regarding the abuse of market dominance.

Chris Phillips, Vice President of Education at Google, stated in an official blog post that schools should have control over AI in their classrooms, and thus Google provided more granular management options. However, because the search function is deeply integrated into its AI architecture, completely disabling AI simultaneously cuts off the search entry point. Previously, Google had opened its AI chatbot Gemini to US students using Google Classroom in August, a move that was widely questioned by parents and educators due to concerns about minors' data security and cognitive development.

The policy announcement quickly drew criticism from regulators and children's rights organizations. Lina Khan, former Chair of the US Federal Trade Commission (FTC), pointed out on social media platform X that Google is using its search monopoly to coerce schools into adopting AI. She emphasized that multiple courts have found Google liable for violating antitrust laws, and this latest action highlights the danger of allowing an unregulated monopolist to control essential public service infrastructure. Meanwhile, cities like New York and Los Angeles had previously attempted to restrict the use of generative AI in elementary and middle schools to assess potential impacts on children's brain development.

Fuentes:spiegel.de

Investigación

Envision Energy Claims World's First Commercial 800V DC Power System Deployment for AI Data Center in China

Material
2026-11-03 20:40 GMT+8

Chinese energy technology company Envision Energy has announced the completion of the world's first commercial deployment of an integrated 800-volt direct current (DC) power system at its Galaxy AI Data Center in Ulanqab, Inner Mongolia. The system, based on the Dubhe Controller and supported by the Dubhe Energy Foundation Model, began testing in August this year and is now operational.

While traditional data centers rely on multiple power-conversion stages to achieve usable supply from the grid, Envision's Dubhe technology converts power directly from 10kV medium-voltage AC to an 800V DC bus. This architecture simplifies the power conversion apparatus, significantly reducing its size and boosting efficiency. The system integrates solid-state transformers (SST), battery storage, and millisecond-level coordination to absorb GPU load fluctuations and provide backup power.

According to the company, the solution supports higher-density AI racks, allowing more computing power in the same area while reducing the requirement for copper cabling. The Galaxy campus plans to scale to 3GW capacity, eventually "supporting up to one million AI accelerators." Furthermore, Envision intends to roll out this 800V DC technology at its Mission Gobi initiative, targeting 5GW of green AI data center capacity in desert regions by 2030.

Fuentes:Tom's Hardware

Investigación
Siguiente →
Iniciar sesión / Registrarse
LecturaInvestigaciónRadarMarco de inversión

Overall, Apple's strategic logic is: with over 2.5 billion active devices, it possesses the distribution interface for AI services; it does not need to build every model itself, nor invest in infrastructure at the scale of its competitors, but can still generate revenue through device sales, platform commissions, and usage-based tiers for its own services. Whether this strategy can sustain depends on two factors—whether customers continue purchasing AI services through Apple's payment system, and whether on-device computing can long-term handle a sufficient proportion of AI inference workloads to maintain low capital expenditure.