Taiwan Financial Supervisory Commission Says it is Drafting ICO Regulations 2157

The Financial Supervisory Commission (FSC) is drawing up national standards for initial coin offerings (ICOs) to make virtual tokens as easy to invest in as stocks and just as liquid, FSC Chairman Wellington Koo said yesterday.

The draft is to be completed by June next year, he told a meeting of the Legislative Yuan’s Finance Committee. Koo made the remarks after Chinese Nationalist Party (KMT) Legislator William Tseng said at a question-and-answer session on whether the government would regulate ICOs.

To conduct an ICO, businesses typically post a white paper on the Internet that outlines their ideas on token use, issue requirements and expected growth before offering virtual tokens for sale.

However, 127 white papers worldwide were found last year to have been faked and as of April, 80 papers were found to be inaccurate, Tseng said. Citing data from ICO advisory firm Satis Group, Tseng said that 81 percent of ICOs have been identified as scams.

The commission would regulate ICOs, Koo said, but added that tokens exchanged for goods, such as those used in accruing points at convenience stores or mileage points accepted by airlines, would not be covered by the standards.

“The more we regulate, the more this new economic behavior wanes,” Koo said.

The commission has no intention of curbing the creativity and productivity associated with cryptocurrencies if they are not used as securities, he said.

“People often confuse an ICO with the trading of cryptocurrencies,” Securities and Futures Bureau Deputy Director-General Tsai Li-ling said by telephone.

An ICO is a type of public fundraising activity, whereas cryptocurrency trading is similar to trading in gold, for which the commission only implements money laundering controls, Tsai said.

If the token in an ICO functioned similar to a security, the commission would define it as a “securities token” and subject it to the Securities and Exchange Act, Tsai said. The issuer would also need to disclose information similar to what companies that are publicly traded need to do now, she said. However, without referencing the standards of other countries, it would be difficult to draft standards for ICOs, Tsai said.

The government tends to regard cryptocurrencies as virtue commodities or assets rather than currencies, because they have no intrinsic value, central bank Governor Yang Chin-long told the committee, explaining why the commission should be responsible for regulating ICOs. The commission said it has helped prosecutors investigate three cases of suspected ICO fraud.

One of the cases involved a private company advertising that coins could be exchanged for equities, which might have violated the Securities and Exchange Act and the Banking Act, the commission said.

Previous ArticleNext Article

Leave a Reply

BrandJet AI Launches Artemis MCP and Introduces Forward Deployed AE Role for AI-Driven GTM Teams 3921

BrandJet AI, a brand intelligence and outreach automation platform, today announced the launch of Artemis, a new Model Context Protocol (MCP) layer designed to help go-to-market (GTM) teams execute complex multi-step workflows using natural language prompts. The company also introduced a new commercial role, the Forward Deployed Account Executive (FDAE), created to support organizations adopting AI-native revenue operations.

The announcements reflect BrandJet AI’s continued focus on reducing fragmentation across sales, marketing, and revenue technology stacks by connecting intent detection and outreach execution within a single operating environment.

Addressing GTM Fragmentation

Revenue teams typically rely on multiple systems to monitor brand conversations, identify prospects, enrich contact data, sequence outreach, and track engagement. These processes often require manual coordination across platforms, creating delays between signal detection and commercial action.

Artemis is designed to streamline this workflow. Built on a Model Context Protocol architecture, it connects BrandJet AI’s monitoring, enrichment, sequencing, and performance-tracking capabilities into a unified prompt-driven layer.

Through Artemis, revenue operators can initiate structured workflows using natural language instructions. For example, a user may request the identification of professionals discussing specific topics across digital platforms within a defined timeframe, enrichment of those profiles, and the creation of an outreach sequence aligned to campaign goals. Artemis coordinates those tasks within the system, allowing teams to reduce operational handoffs.

According to BrandJet AI, the goal is not to replace strategic oversight but to simplify execution.

“Revenue teams spend too much time stitching together tools instead of acting on real buying signals,” said Nirav Shah, CEO of BrandJet AI. “Artemis helps unify intelligence and execution so teams can move from insight to outreach more efficiently.”

Prompt-Driven Workflow Orchestration

Artemis supports workflows that include:

  • Monitoring brand and competitor mentions across social platforms and the open web
  • Identifying potential prospects based on observable intent signals
  • Enriching lead data within the platform
  • Initiating multi-channel outreach across email and major social networks
  • Tracking engagement and campaign performance in real time

Rather than requiring operators to manually transfer data between systems, Artemis enables coordinated execution through a conversational interface layered on top of BrandJet AI’s infrastructure.

The system is designed to operate within compliance and governance standards established by customer organizations, maintaining human oversight over messaging and campaign parameters.

Introducing the Forward Deployed Account Executive

Alongside the Artemis launch, BrandJet AI announced the introduction of the Forward Deployed Account Executive (FDAE), a role intended to help enterprise customers integrate AI-driven workflows into their revenue operations.

As AI platforms become more advanced, organizations often encounter implementation gaps between technical capability and day-to-day usage. The FDAE model is structured to address that gap by embedding commercially accountable operators more deeply into customer environments.

Unlike traditional account executives who primarily focus on closing new business, or customer success managers who focus on support and retention, the FDAE combines revenue accountability with workflow strategy support. The role is designed to assist customers in mapping Artemis and broader BrandJet AI capabilities to their specific GTM structures.

“The technology layer is evolving quickly, but successful adoption depends on workflow design and operational alignment,” said Marsad Aurangzeb, Founder of BrandJet AI. “The Forward Deployed AE role is intended to help customers translate AI capabilities into measurable revenue outcomes.”

BrandJet AI plans to formalize the FDAE framework and publish additional details regarding the role’s structure and responsibilities in 2026.

Connecting Listening and Outreach

Historically, social listening and sales engagement technologies have evolved separately. Listening platforms track conversations, brand mentions, and sentiment across digital channels, while engagement platforms focus on outbound sequencing and pipeline development.

BrandJet AI’s platform integrates both functions, allowing teams to identify signals and initiate outreach within the same environment. With Artemis, those processes can now be coordinated through prompt-driven workflows.

For example, when a relevant public conversation surfaces online, such as a discussion about a specific technology category, hiring signals, or operational challenges, Artemis can help surface the signal, enrich the associated contact, and assist in preparing a tailored outreach campaign.

The objective is to reduce the time between observed intent and commercial response, while maintaining alignment with compliance and messaging standards.

Enterprise Implementation and Governance

BrandJet AI emphasizes that Artemis is built to operate within enterprise governance frameworks. Campaign parameters, messaging templates, and data usage policies remain configurable by customer teams.

As organizations expand AI adoption within revenue functions, governance considerations, including messaging accuracy, compliance adherence, and brand alignment, remain central. Artemis is positioned as an execution layer that operates within these controls rather than outside them.

The company states that ongoing enhancements are planned, including additional intent modeling refinements, deeper workflow customization, and validation loops that compare forecasted campaign outcomes with actual engagement performance over time.

Availability

Artemis MCP is available immediately to customers on BrandJet AI’s Growth and Enterprise plans. Availability for Starter plan users is expected in Q2 2026. Forward Deployed Account Executive engagements are currently offered on a limited basis for Enterprise customers.

Organizations interested in learning more may contact BrandJet AI directly for additional information.

About BrandJet AI

BrandJet AI is a brand intelligence and outreach automation platform designed for modern revenue teams. The platform enables organizations to monitor brand and competitor activity across digital channels, identify potential prospects based on social and behavioral signals, and execute multi-channel outreach campaigns within a unified interface. BrandJet AI serves growth-stage and enterprise organizations across SaaS, financial services, and professional services industries.

Xojo 2025r3 Delivers Major Updates, New Libraries Feature, Integrated AI Assistant, and Modern OS Support 4549

Xojo, Inc., the developers behind Xojo—a powerful cross-platform development tool and programming language— announce the immediate availability of Xojo 2025 Release 3, a significant update that expands platform compatibility, introduces powerful new development tools and enhances performance across Desktop, Web, Console, iOS and Android.

This release adds official support for macOS Tahoe 26 and iOS 26, including Apple’s new Liquid Glass interface technologies, ensuring developers can confidently build and deploy apps on the latest operating systems.

Xojo 2025r3 debuts Libraries, a major enhancement to the IDE that allows developers to package and reuse compiled classes and interface elements across Desktop, Web, Console and iOS projects. Libraries make sharing and versioning custom functionality easier and faster.

The IDE also introduces Jade, Xojo’s new integrated AI assistant. Jade helps developers write code more efficiently, suggest improvements and accelerate learning directly from inside the Xojo environment. “Xojo 2025r3 is one of our most forward-looking releases yet. With Libraries, modern platform support and our new AI assistant Jade, we’re giving developers powerful tools that help them work faster, build smarter and deliver great apps on every platform,” says Geoff Perlman, Xojo’s Founder and CEO.

New Features and Updates:

  • Support for macOS Tahoe 26 and iOS 26
  • IDE now supports Libraries for Desktop, Web, Console and iOS projects
  • IDE now has an AI assistant, called Jade
  • Added DesktopGrid control
  • Multiple WebListBox improvements
  • Web now uses Bootstrap v5.3.7 and Bootstrap Icons v1.13.1
  • Added Passkeys/WebAuth support for web apps
  • Windows DesktopXAMLContainer improvements
  • Expanded WinUI-backed controls
  • Several Crypto enhancements
  • Improved color settings for layouts and controls on iOS and Android
  • Android Support for PDFDocument
  • Android now uses Kotlin 2.2.20 and targets Android 16 (SDK 36)

About Xojo

Xojo is a cross-platform development tool for building native apps for macOS, Windows, Linux, iOS, Android, the web and Raspberry Pi. For over 25 years, Xojo has supported a growing community of developers passionate about creating powerful applications with ease. Learn more at xojo.com. Download Xojo 2025 Release 3 at xojo.com/download.

Availability

Xojo is free for learning and development, as well as for building apps for Linux and Raspberry Pi. Paid licenses start at $499 for cross-platform Desktop, Mobile, or Web development. Xojo Pro and Pro Plus licenses, starting at $999, offer additional support and resources for professional developers. Special licensing is available for educators and students. Visit xojo.com/store for details.

AI Infrastructure Company EverMind Released EverMemOS, Responding to Profound Challenges in AI 4657

AI infrastructure company EverMind has recently released EverMemOS, an open-source Memory Operating System designed to address one of artificial intelligence’s most profound challenges: equipping machines with scalable, long-term memory.

The Memory Bottleneck

For years, large language models (LLMs) have been constrained by fixed context windows, a limitation that causes “forgetfulness” in long-term tasks. This results in broken context, factual inconsistencies, and an inability to deliver deep personalization or maintain knowledge coherence. The issue extends beyond technical hurdles; it represents an evolutionary bottleneck for AI. An entity without memory cannot exhibit behavioral consistency or initiative, let alone achieve self-evolution. Personalization, consistency, and proactivity, which are considered the hallmarks of intelligence, all depend on a robust memory system.

There is a consensus that memory is becoming the core competitive edge and defining boundary of future AI. Yet existing solutions, such as Retrieval-Augmented Generation (RAG) and fragmented memory systems, remain limited in scope, failing to support both 1-on-1 companion use cases and complex multi-agent enterprise collaboration. Few meet the standard of precision, speed, usability, and adaptability required for widespread adoption. Equipping large models with a high-performance, pluggable memory module remains a core unmet demand across AI applications.

Discoverative Intelligence

“Discoverative Intelligence” is a concept proposed in late 2025 by entrepreneur and philanthropist Chen Tianqiao. Unlike generative AI, which mimics human output by processing existing data, Discoverative Intelligence describes an advanced AI form that actively asks questions, forms testable hypotheses, and discovers new scientific principles. It prioritizes understanding causality and underlying principles over statistical patterns, a shift Chen argues is essential to achieving Artificial General Intelligence (AGI).

Chen contrasted two dominant AI development paths: the “Scaling Path,” which relies on expanding parameters, data, and compute power to extrapolate within a search space, and the “Structural Path,” which focuses on the “cognitive anatomy” of intelligence and how systems operate over time.

Discoverative Intelligence falls into the latter category, built on a brain-inspired model called Structured Temporal Intelligence (STI) that requires five core capabilities in a closed loop: neural dynamics (sustained, self-organizing activity to keep systems “alive”), long-term memory (storing and selectively forgetting experiences to build knowledge), causal reasoning (inferring “why” events occur), world modeling (an internal simulation of reality for prediction), and metacognition & intrinsic motivation (curiosity-driven exploration, not just external rewards).

Among these capabilities, long-term memory serves as the vital link between time and intelligence, highlighting its indispensable role in the path toward achieving true AGI.

EverMind’s Answer

EverMemOS is EverMind’s answer to this need: an open-source Memory Operating System designed as foundational technology for Discoverative Intelligence. Inspired by the hierarchical organization of the human memory system, EverMemOS features a four-layer architecture analogous to key brain regions: an Agentic Layer (task planning, mirroring the prefrontal cortex), a Memory Layer (long-term storage, like cortical networks), an Index Layer (associative retrieval, drawing from the hippocampus), and an API/MCP Interface Layer (external integration, serving as AI’s “sensory interface”).

The system delivers breakthroughs in both scenario coverage and technical performance. It is the first memory system capable of supporting both 1-on-1 conversation use cases and complex multi-agent enterprise collaboration. On technical benchmarks, EverMemOS achieved 92.3% accuracy on LoCoMo (a long-context memory evaluation) and 82% on LongMemEval-S (a suite for assessing long-term memory retention), significantly surpassing prior state-of-the-art results and setting a new industry standard.

The open-source version of EverMemOS is now available on GitHub, with a cloud service version to be launched late this year. The dual-track model, combining open collaboration with managed cloud services, aims to drive industry-wide evolution in long-term memory technology, inviting developers, enterprises, and researchers to contribute to and benefit from the system.

About EverMind

EverMind is redefining the future of AI by solving one of its most fundamental limitations: long-term memory. Its flagship platform, EverMemOS, introduces a breakthrough architecture for scalable and customizable memory systems, enabling AI to operate with extended context, maintain behavioral consistency, and improve through continuous interaction.

To learn more about EverMind and EverMemOS, please visit:
Website: https://evermind.ai/
GitHub: https://github.com/EverMind-AI/EverMemOS
X: https://x.com/EverMindAI
Reddit: https://www.reddit.com/r/EverMindAI/

Salt Security Brings MCP Threat Protection to AWS WAF, Blocking AI Agent Abuse in Real Time 4244

Salt Security, the leader in API security, today announced it is extending its patented, award-winning API behavioral threat protection to detect and block malicious intent targeting Model Context Protocol (MCP) servers deployed within the AWS ecosystem. Building on the recent launch of Salt’s MCP Finder technology, Salt now enables organizations to identify external misuse and abuse of MCP servers by AI agents and attackers, and automatically block these threats using its integration with AWS WAF.

MCP servers have rapidly become a key component of enterprise AI architecture, enabling LLMs and autonomous agents to call APIs, execute tools, and complete workflows. But they also represent a new threat vector. Deployed without central oversight and often exposed to the internet, MCP servers are increasingly targeted by adversaries for unauthorized access to critical data and system access.

With this new capability, Salt enables customers to use their existing AWS WAF deployments to block attacks on MCP infrastructure. The protections are informed by real-time behavioral threat data from Salt’s platform.

“Most organizations don’t even know how many MCP servers they have, let alone which ones are exposed or being abused,” said Nick Rago, VP of Product Strategy at Salt Security. “This capability lets them take action quickly, using existing controls to prevent real threats without needing to deploy new infrastructure.”

The solution is based on Salt’s MCP Finder technology, which provides full visibility into the MCP layer across external, internal, and shadow deployments. By combining that discovery with AWS WAF, customers can:

  • Automatically block MCP misuse and abuse before it impacts applications
  • Discover previously unknown or unmanaged MCP implementations and ensure traffic is routed through AWS WAF for inspection and protection
  • Extend AWS WAF edge protection to the AI action layer
  • Apply intent-based behavioral threat detection to stop attacks targeting key AI infrastructure that traditional tools miss
  • Continuously update protections based on evolving attacker tactics

Salt Security is showcasing these capabilities at AWS re:Invent 2025. The integration is available now as part of the Salt Security API Protection Platform.

About Salt Security

Salt Security secures the APIs that power today’s digital businesses. Salt delivers the fastest API discovery in the industry—surfacing shadow, zombie, and unknown APIs before attackers find them. The company’s posture governance engine and centralized Policy Hub automate security checks and enforce safe API development at scale. With built-in rules and customizable policies, Salt makes it easy to stay ahead of compliance and reduce API risk. Salt also uses machine learning and AI to detect threats early, giving companies a critical advantage against today’s sophisticated API attacks. The world’s leading organizations trust Salt to find API gaps fast, shut down risks, and keep their businesses moving. Learn more at https://salt.security

NeuralTrust introduces Guardian Agents: the first AI agents built to protect other agents 3226

NeuralTrust, the security platform for AI Agents and LLM applications, today announced Guardian Agents, a new class of autonomous security agents designed to defend enterprise AI systems in real time. As organizations deploy thousands of AI agents, each connected to tools, APIs, and sensitive workflows, Guardian Agents provide a dedicated, agent-native layer of protection.

Unlike traditional security controls built for static applications, Guardian Agents are active defenders. They monitor agent behavior, intercept unsafe actions, enforce tool-use policies, scan for vulnerabilities, and stop attacks before they escalate.

A new force to counter a new threat landscape

Enterprises today face an unprecedented operational challenge. AI agents can write code, move data, trigger workflows, and interact with external systems. At scale, the risk surface becomes ungovernable:

  • A single agent may access hundreds of tools
  • One misconfigured workflow can leak sensitive data
  • A prompt injection can escalate privileges or bypass guardrails

Guardian Agents act as a protective layer around this ecosystem. Instead of relying solely on static filters or manual governance, NeuralTrust gives security teams their own force of autonomous defenders to act at machine speed.

How Guardian Agents work

Rather than blocking innovation, Guardian Agents sit alongside production agents to ensure safe execution. They:

  • Stop complex attacks such as prompt injection, privilege escalation, and malicious tool use
  • Prevent data leaks by inspecting inputs, outputs, and tool interactions
  • Enforce granular policies defining exactly which tools, actions, and permissions each agent can use
  • Scan AI applications to uncover vulnerabilities, unsafe flows, and misconfigurations
  • Analyze behavior to detect anomalies and emerging threats
  • Leverage a continuously updated threat database engineered specifically for AI agents

Guardian Agents are deployed through NeuralTrust’s high-performance security platform, which processes billions of requests every month. Purpose-built for LLM and agent workloads, it delivers industry-leading performance with minimal latency, and works across all clouds, models, and integrations.

“Autonomous agents have changed the threat landscape. Defending them requires security that moves just as fast,” said Joan Vendrell, Co-Founder and CEO of NeuralTrust. “Guardian Agents give organizations a way to stay ahead of attacks, enforce policy, and deploy AI safely at scale.”

About NeuralTrust

NeuralTrust is the leading platform for securing and scaling AI Agents and LLM applications. Recognized by the European Commission as a champion in AI security, we partner with global enterprises to protect their most critical AI systems. Our technology detects vulnerabilities, hallucinations, and hidden risks before they cause damage, empowering teams to deploy AI with confidence.

Learn more at neuraltrust.ai.

Allstate launches Scam Protection to safeguard workers’ finances from rising cyber and crypto fraud 2430

  • Allstate launches Scam Protection, a new workplace benefit that covers modern digital threats including scams, ransomware and cryptocurrency theft, featuring first-of-its-kind reimbursement up to $50,000 for verified scam losses.
  • Employers turn to Allstate amid record-breaking cybercrime trends. The insurer saved customers $33.2 million last year in potential losses, with holiday shopping periods like Black Friday and Cyber Monday driving the steep spikes in fraud claims.
  • Allstate Scam Protection is available to nearly 7 million people and their families through employer benefits. Companies are adding it to benefits packages to safeguard employees’ personal finances during open enrollment.
  • Coverage extends beyond employees to protect family members, including aging parents 65 and older, regardless of where they reside.

Allstate is helping employers safeguard workers’ finances and fight cybercrime with Allstate Scam Protection, including a first-of-its-kind reimbursement up to $50,000 when money, including cryptocurrency, is stolen through scams. The coverage is available exclusively through workplace benefits, reaching nearly 7 million employees and their families during open enrollment. Employees can check their workplace benefits or ask their HR team to see if Allstate Scam Protection is offered and how to enroll.

Cybercrime is surging, straining household finances and workplace productivity. Allstate Identity Protection recovered $33.2 million in potential identity theft losses for customers in 2024, with the fourth quarter alone costing $9.8 million. New account fraud drove $23.3 million in actual losses last year, while fraudulent applications added $5.5 million in potential exposure. Each identity theft case takes time and resources to resolve, draining employee time and productivity that often overlaps with the workday.

This new protection comes as cybercrime spikes during the holiday shopping season, especially around Black Friday and Cyber Monday, when phishing texts, fake retailers and social media scams are rampant.

“Scams cost employees time, money and productivity, impacting their families and their finances,” said Caroline Slane, senior vice president of business operations at Allstate Identity Protection. “This new scam protection fills a gap by putting money back in people’s hands so they can get back on their feet.”

What does Allstate Scam Protection cover?

Allstate Scam Protection goes beyond traditional identity theft products by covering new ways criminals steal money online with fewer exclusions. Here’s what’s included:

  • Reimbursement for scams, digital crimes and social engineering: Includes unlimited claims up to $50,000 per year.
  • Cryptocurrency theft coverage: Reimburses stolen crypto resulting from cybercrime up to $50,000 per year.
  • Web, email and mobile protection: Ensures safe web browsing, protects and alerts against fraudulent texts, emails, links, robocalls and robotexts.
  • Family coverage: The benefit covers unlimited household members, including teens and seniors 65 and older, providing protection for those most vulnerable, even if they live outside the home.
  • Scam takedown: Customers can report malicious URLs directly to Allstate’s cyber experts for removal.
  • Personal coaching: Individual sessions with Allstate Identity Protection specialists to build tailored defense plans for individuals and families.

Why are employers adding Allstate Scam Protection to benefits packages?

Allstate Identity Protection products are already widely available through employee benefits packages at 4,500 companies including a quarter of the Fortune 500. Employers are adding Allstate Scam Protection because:

  • Fraud losses reached $12.5 billion in 2024, according to the FTC. While the number of fraud reports stayed flat, far more people lost money compared to the year prior.
  • AI has supercharged scams. AI fuels phishing emails, impersonation texts and deepfake voices or videos that are nearly impossible to spot.
  • Cybercrime impacts workplace productivity. Workers are vulnerable whenever and wherever they use devices, and they manage fraud recovery during work hours when banks and government agencies are open.

How can consumers protect themselves during holiday shopping season?

Holiday scams surge around Black Friday and Cyber Monday. Allstate recommends:

  • Slow down before buying. Beware of hard-to-find items on third-party sites and avoid sellers demanding gift card payments.
  • Protect your wallet and identity while shopping online. Pay with credit cards, keep track of purchases, update device software and avoid clicking links in unfamiliar texts or emails.
  • Use digital wallets in stores. They encrypt payment info and reduce the chance of card skimming or data theft.

About Allstate

The Allstate Corporation protects people from life’s uncertainties with affordable, simple and connected protection for autos, homes, electronic devices, and identities. Products are available through a broad distribution network including Allstate agents, independent agents, major retailers, online, and at the workplace. Allstate has more than 209 million policies in force and is widely known for the slogan “You’re in Good Hands with Allstate.” For more information, visit www.allstate.com.