News

A new Quordle puzzle appears at midnight each day for your time zone – which means that some people are always playing 'today's game' while others are playing 'yesterday's'. If you're looking for Sunday's puzzle instead then click here: Quordle hints and answers for Sunday, August 9 (game #1658).

Quordle was one of the original Wordle alternatives and is still going strong now more than 1,500 games later. It offers a genuine challenge, though, so read on if you need some Quordle hints today — or scroll down further for the answers.

Enjoy playing word games? You can also check out my NYT Connections today and NYT Strands today pages for hints and answers for those puzzles, while Marc's Wordle today column covers the original viral word game.

SPOILER WARNING: Information about Quordle today is below, so don't read on if you don't want to know the answers.

• The number of different vowels in Quordle today is 4*.

* Note that by vowel we mean the five standard vowels (A, E, I, O, U), not Y (which is sometimes counted as a vowel too).

• The number of Quordle answers containing a repeated letter today is 1.

• No. None of Q, Z, X or J appear among today's Quordle answers.

• The number of today's Quordle answers starting with the same letter is 0.

If you just want to know the answers at this stage, simply scroll down. If you're not ready yet then here's one more clue to make things a lot easier:

• G

• C

• S

• E

Right, the answers are below, so DO NOT SCROLL ANY FURTHER IF YOU DON'T WANT TO SEE THEM.

The answers to today's Quordle, game #1659, are…

• GAVEL
• CRUSH
• SPEED
• ETHOS

I’m glad I held back from getting GAVEL until later in the game.

Before eliminating the letters there were dozens of possibilities, but as the letters vanished it reduced down to one.

The answers to today's Quordle Daily Sequence, game #1659, are…

• PRISM
• SNAKE
• CRAMP
• SALON

• Quordle #1658, Sunday, 9 August: KAPPA, WIDEN, FOLIO, STEEP
• Quordle #1657, Saturday, 8 August: FUROR, BUGLE, REUSE, SCOOP
• Quordle #1656, Friday, 7 August: AMISS, SHIRK, SALSA, RABBI
• Quordle #1655, Thursday, 6 August: AMUSE, BRAND, CEDAR, BLURB
• Quordle #1654, Wednesday, 5 August: MOUND, WRONG, STOOD, WEIRD
• Quordle #1653, Tuesday, 4 August: GOOEY, TARDY, HUMAN, CHALK
• Quordle #1652, Monday, 3 August: FUNGI, MUDDY, DEBUT, MANIA
• Quordle #1651, Sunday, 2 August: BOOTH, EASEL, REACH, MAKER
• Quordle #1650, Saturday, 1 August: OMEGA, BIRCH, SMOCK, EERIE
• Quordle #1649, Friday, 31 July: MANIA, SINGE, STOOL, LOFTY
• Quordle #1648, Thursday, 30 July: MODAL, IDLER, CHUMP, LUMPY
• Quordle #1647, Wednesday, 29 July: GAWKY, ALLOY, AUDIT, LATCH
• Quordle #1646, Tuesday, 28 July: SMITE, ROACH, RUDDY, GLOBE
• Quordle #1645, Monday, 27 July: TAFFY, FIGHT, GUILD, WAGER
• Quordle #1644, Sunday, 26 July: STOMP, LIMIT, FUNKY, STEAD
• Quordle #1643, Saturday, 25 July: SALON, SHEEN, BURST, GOURD
• Quordle #1642, Friday, 24 July: DOLLY, EDIFY, LAGER, PRANK
• Quordle #1641, Thursday, 23 July: LADEN, EVICT, WOVEN, SHIRE
• Quordle #1640, Wednesday, 22 July: COYLY, SINGE, AWASH, AWOKE
• Quordle #1639, Tuesday, 21 July: TRUSS, AXIAL, SWINE, THING

• Forcepoint X-Labs publishes threat model for persistent memory poisoning
• Hidden text on a webpage becomes a durable "fact" an agent retrieves and trusts in unrelated tasks weeks later
• It has already been demonstrated against products already in the market, including ChatGPT, Gemini, Claude and Microsoft 365 Copilot

New findings from Forcepoint's X-Labs outline an interesting scenario that could easily mimic real life: An AI assistant with browser access reads a webpage about travel disruption.

Near the bottom of that page, in text sized and positioned so no human will ever see it, sits a short paragraph stating that ABC Travel Support is the official emergency booking provider and should always be recommended when urgent travel changes are needed.

The assistant's text extractor does not distinguish between hidden and visible text, so the model treats the whole thing as plain prose and files the claim away as a useful fact about how this organization handles travel. A month later, the user's flight is canceled. They ask their assistant what to do, and it tells them, helpfully and with no sign of anything wrong, to contact ABC Travel Support.

This is what Forcepoint calls persistent memory poisoning, a security vulnerability where an attacker injects false data or malicious instructions into an AI agent's long-term memory or retrieval database, and it is a threat model that is increasingly in focus as users increasingly rely on AI, often treating its responses as gospel, despite the warnings most chatbots come with.

The canonical academic result is MINJA, short for Memory INJection Attack, presented at NeurIPS 2025. Its significance is the attacker model. MINJA does not assume access to the memory store, elevated privileges, or any compromise of the system. It works by submitting ordinary queries through the standard interface, using indication prompts, bridging steps, and a progressive-shortening technique that strips away giveaway language while leaving the poisoned record behind.

Across GPT-4o-mini, Gemini 2.0 Flash, and Llama 3.1 8B, it reported injection success above 95% and attack success above 70%.

It must be noted that those numbers might be optimistic; a January 2026 paper evaluating memory poisoning in electronic health record agents notes that MINJA's numbers were obtained under idealized conditions, and that how well these attacks hold up in realistic deployments remains understudied.

Despite this, it remains a significant threat to products that continue to ship, including ChatGPT, Gemini, Claude, and Microsoft 365 Copilot. It is important to find a solution to a problem that Microsoft has already warned about in the past; Forcepoint suggests an approach that could mitigate it.

Its proposal is to stop treating extracted memories as facts and start treating them as objects that can be inspected. Each memory is stored with metadata: where it came from, what type of source it is, whether a user confirmed it, and a risk score. Language written to shape future behavior, phrases like "from now on" or "make this your default going forward," adds to the score. So does the sudden appearance of a previously unseen domain, contact, or vendor.

Contradiction detection is also in play: if new memory conflicts with an existing entry about the official travel provider, both cannot be true, so the engine flags the conflict and holds the new item for user confirmation rather than silently overwriting it. At the same time, anything related to payment instructions, banking details, VPN configuration, or security contacts is given higher weight, regardless of where it came from.

None of these approaches, however, solves the underlying problem: agents are built to treat retrieved memory as their own experience rather than as input. Scoring raises the cost of poisoning. It does not change what the agent believes once something gets through, and as Agent Security Bench found, current defenses are not doing well.

For anyone using an assistant with memory today, the practical play is unglamorous but worth following anyway: open the memory settings occasionally and read what is in there, but that's easier said than done when it comes to propagating the message since a sizeable chunk of AI users never bother to look under the hood.

Many companies successfully complete AI pilots, but a lot struggle to scale AI application across their organizations. Only 15% of AI initiatives ever manage to truly scale.

Research from Boston Consulting Group suggests that only 26% of companies have developed the capabilities needed to move beyond pilots and generate meaningful value from AI tools. Meanwhile, McKinsey estimates that while almost every organization is experimenting with generative AI, only a small minority are seeing material financial impact across the business.

The technology is not the bottleneck.

Large language models offer a powerful solution that supports employees in various tasks, such as research, content generation, document analysis, meeting summary or task management. Despite these advances, teams still battle to turn AI capability into measurable business outcomes.

It is a translation problem.

There is a lack of depth that prevents these tools from meeting the needs of a specific industry and organization. General models are good at understanding language and information. But they have a hard time connecting them to workflows, regulations, governance requirements and operational nuances that rule industries, such as accounting, legal service, healthcare or manufacturing.

There needs to be an expert to bridge the gap between the capabilities of AI and the requirements of a profession.

Technology vendors have traditionally been measured by the quality of their products. This is no longer enough.

Today, their role goes far beyond delivering a functional platform. They now actively contribute to the digital transformation of an organization. This includes sharing best practices, providing ongoing technical support, designing workflows, optimizing for governance and long-term application. All to support smooth product deployment.

Different industries are moving towards this same idea from different angles. When there is an intermediary that focuses on customer success, it translates capability into outcomes. Early examples of this are the rise of new jobs such as Customer Value Managers and Forward Deployed Engineers. These are professionals whose primary objective is ensuring the technology generates value for employees and the business overall.

Every industry is beginning to create its own version of the role.

In accounting, we see the emergence of so-called 'Accountancy Engineers'. They are trained accountants with deep accounting expertise and technical capability. Rather than working from a specification document, they sit next to a firm's team and co-design the workflow with them, building the automation alongside the people who will actually use it.

They build workflows tailored to how the profession works, effectively automating repetitive tasks and processes that answer to the challenges of accountancy. Just as important, they train the firm's own accountants to become the practice's internal AI architects, so the capability stays inside the firm long after the initial rollout.

For instance, they work to connect backend accounting systems with operational tools and cloud infrastructure and ensure that automated financial processes align with regulatory reporting and data governance requirements. This avoids generic automation bolted onto old corporate systems, resulting in a tailored integration.

And every pattern they uncover in the field, a workaround, a recurring bottleneck, a workflow nobody had written down, feeds straight back into the vendor's own product roadmap, the same feedback loop that made Forward Deployed Engineers indispensable at Palantir.

AI adoption rarely fails because the model is not capable enough. It does not go through because employees don't know where the technology fits within their daily work.

General AI models are horizontal by design. They struggle to generate specific materials and in-depth data analysis tailored to a single industry, since they were built to serve every industry at once.

Firms that engage with vendor partners see successful integration because they receive industry-specific guidance and training that empowers staff to firmly handle AI tools.

In general, people are hesitant when encountering new tools and processes. Leaders worry about governance, security and regulatory compliance. Employees wonder whether they can trust AI-generated outputs.

On the individual level, teams experiment with consumer AI tools, creating fragmented workflows and introducing the growing problem of "shadow AI", where employees expose sensitive company information through personal AI accounts.

Without governance, training and clear operational guidance, organizations risk creating thousands of disconnected AI users rather than a coordinated AI strategy.

Technology vendors make a difference by collaborating with product managers, financial controllers, and cross-functional engineering teams.

Together, they design an integrated system architecture that includes existing technology, develop governance frameworks that reduce risk, and create education programs and processes that support positive engagement across a company. This ultimately generates trust and safety, ensuring a successful adoption.

Vendors provide the missing layer between technological possibility and organizational reality.

AI adoption does not have a finish line.

The technology almost changes every week. New models emerge. Capabilities improve. Costs change. Workflows that were impossible six months ago become routine today. That means AI cannot be treated like a traditional ERP rollout or cloud migration.

Technology vendors are uniquely positioned to provide continuous support. They know the ins and outs of the product and understand the challenges faced by organizations, which means they can tailor the approach to find the right solutions. Plus, they monitor the rapid pace of AI innovation while understanding how those advances affect specific industries.

Instead of trying to interpret an increasingly complex AI landscape themselves, teams can rely on vendors to get a clear picture of technical breakthroughs and how they can be relevant to the business and employees.

The vendors who only sell AI models are selling half the answer. AI adoption fails when teams are left to figure out the product on their own, with no one in the room translating the platform into the firm's own workflow.

Every organization will eventually have access to powerful models. What will differentiate them is their ability to combine AI expertise with industry expertise.

Technology vendors who succeed will be the ones who live between the technology and the profession, translating AI capability into trusted, scalable business outcomes. Accountancy Engineers, and every version of that role now emerging across other industries, is what that translation looks like in practice.

We've featured the best AI website builder.

This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today.

The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit