
Adapt or else? How the sell-side is adjusting to new risks
There is also a trade-off to be considered – which other compliance project must be deprioritized to provide financial headroom for any new rules? 'Uncertainty of when, or if, a regulation is required adds a risk to a bank, who have to make a decision on how to invest engineering resources to meet regulations,' says Allright.
Technology risks
Technology itself is adding risks to banking operations. As Bloomberg's Global Head of Sell-Side Product Phil McCabe explained earlier in this report, the advent of electronic trading and automation has compressed investors' fees, hastening the diversification of their portfolios into new markets as they search of better returns. This has brought liquidity to once niche asset classes such as crypto, private equity and structured products
Moving into new markets requires investment to develop new complex pricing models and deploy techniques to clean and smooth data. These costs can increase further when there is a requirement to provide portfolio level analytics in a timely manner.
Search for innovative solutions
Focusing on data and its increasing significance in banking operations: consistent datasets, particularly when used for pricing financial securities, provide banks with reliable informational bases for assessing and addressing emerging risks. Additionally, APIs facilitate connectivity to software and modeling tools, enabling banks to effectively utilise the expanding data pools accumulated by the sell-side on a daily basis.
However, as the sophistication of data uses grows, many banks are unlikely to have necessary capabilities in-house to smooth curves and clean data sources as well as accumulate all data sets in a timely fashion, and so they increasingly need to turn to data- and technology-specialist vendors.
Mitigation technology
Sell-side organizations must consider how to achieve technological transformation to address new risk profiles. Building capacity is one option, but it is costly and requires ongoing updates. This strategy may be impractical if the organization's tech stack is fragmented, and systems are poorly connected.
As the speed of technological change and the cost of implementation rise, it has become prudent for banks to outsource these operations.
A key to meeting the challenges of the new regulatory landscape is access to consistent data sets and pricing models, says Allright. 'What you [as a bank] don't want to have is a risk solution at the top of the house that's different from a solution that the traders are utilizing, because it's likely they will be looking at different numbers,' he says.
In organisations with siloed tech stacks, end users such as traders and compliance professionals, may not be working from single source of truth dataset, or a golden source of their enterprise intelligence. This becomes challenging when different departments may have to report to different regulators with different data sets.
Having access to consistent data and pricing models ensures banks can satisfy modern reporting rules because regulators require data transparency and accurate pricing models that can justify the claims and assumptions built into disclosures.
An additional reason to seek external support is the ability to connect to data and models in structures such as clouds and APIs, which allows for secure integration of external and internal data and provides organizations with a comprehensive view of their risk exposures.
It is of great importance to banks to bring huge volumes of quality data into their systems ready for analysis and querying, says Allright.
To that end, Bloomberg has developed tools and technology that allows -banks to access software and infrastructure through the cloud and APIs without the need to for large capital outlays. For instance, Bloomberg's Web Services technology delivers lightweight, industry-standard APIs.
These APIs provide banks with flexible access to Bloomberg's software and infrastructure without requiring significant capital investment. 'A lot of investment in the past has gone into hardware on site, but these are machines that are not as flexible and scalable as we move into a new cloud-based computing world,' Allright says.
'We see clients trying to access our set of APIs rather than buy in a fixed product that sits on their desktop because they want to utilize the data and manage the data in different ways increasing deploying their own artificial intelligence – they want to commingle it with their data and create their own IP.'
Hashtags

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
21 minutes ago
- Yahoo
China proposes global cooperation body on AI
SHANGHAI, July 27, 2025 /PRNewswire/ -- A news report from The Chinese government has proposed the establishment of a world AI cooperation organization as part of its efforts to bolster open, inclusive and equitable artificial intelligence development and governance globally. Premier Li Qiang announced the initiative when addressing the opening ceremony of the 2025 World AI Conference and High-Level Meeting on Global AI Governance in Shanghai. Li emphasized the need for collaborative approaches to global AI governance to ensure that intelligent technologies are developed for the good of all. More than 1,000 high-level representatives from upwards of 40 countries and international organizations attended the opening ceremony of the three-day event, themed "Global Solidarity in the AI Era". Li said that the risks and challenges brought by AI development, such as machine hallucinations, deep fakes and AI proliferation, have aroused widespread attention. There is an urgent need to further consolidate a societal consensus on how to achieve a balance between development and security in the AI sector, he said. Li said greater emphasis should be placed on collective governance to ensure that AI ultimately benefits humanity. Given that global AI governance appears fragmented, it is essential to enhance coordination and alignment among countries around the world to establish a framework and rules for global AI governance with a broad consensus, he added. Noting that China places great importance on global AI governance and actively participates in promoting multilateral and bilateral cooperation in this regard, the premier said the nation is willing to offer more Chinese solutions to the international community and contribute more Chinese wisdom to global AI governance. In 2023, China proposed the Global AI Governance Initiative to promote an open, inclusive and fair approach to the development and governance of AI technologies. Last year, China proposed the AI Capacity-Building Action Plan for Good and for All to bridge the AI and digital divides, and to ensure the Global South benefits equitably from AI development. China actively promotes open source development, and is willing to collaborate with countries around the world to promote progress in software and hardware technologies, intensify open source initiatives, and collectively propel AI development to higher levels, Li said. Also on July 26, the conference adopted the Global AI Governance Action Plan, which calls on all parties to work in solidarity to advance AI development and governance. For more information: View original content to download multimedia: SOURCE Sign in to access your portfolio
Yahoo
21 minutes ago
- Yahoo
Elon Musk's SpaceX Moves Bitcoin For The First Time In 3 Years: Is A Sell-Off Incoming?
Benzinga and Yahoo Finance LLC may earn commission or revenue on some items through the links below. SpaceX has reportedly moved parts of its Bitcoin holdings after a long period of dormancy. Cryptocurrency intelligence platform Arkham said Tuesday on X that a wallet linked to the Elon Musk-led space exploration and technology company had moved about 1,300 BTC worth over $153 million, marking the first time the wallet has been touched since June 2022. 'SPACEX JUST MOVED BITCOIN FOR THE FIRST TIME IN 3 YEARS,' Arkham wrote. 'They sent 1.3K BTC ($153M) to a fresh address this morning.' Don't Miss: 7,000+ investors have joined Timeplast's mission to eliminate microplastics— — no wallets, just price speculation and free paper trading to practice different strategies. The report has raised questions about the purpose of the transfer, with moves like this typically indicative of a wallet custody adjustment or an impending sell-off. SpaceX did not immediately respond to a Benzinga request for comment. According to Arkham data, the SpaceX-linked wallet still holds nearly 7,000 BTC worth over $830 million. Meanwhile, at last look, the transferred 1,300 BTC has not moved from the recipient address. The recent asset movement comes as SpaceX's lucrative government contracts have reportedly come under scrutiny from the Trump administration following Musk's row with President Donald Trump. Meanwhile, amid this uncertainty, the firm is seeking to raise over $1 billion to secure a $400 billion valuation. Trending: Grow your IRA or 401(k) with Crypto – . SpaceX's Bitcoin Exploration Musk first disclosed that SpaceX had added Bitcoin to its balance sheet in July 2021, without revealing how much the firm had invested in the asset. But reporting from blockchain sleuths suggests that the firm purchased nearly 26,000 BTC for about $860 million in 2021 at an average price of $33,000 per coin. Sometime down the line, however, the reporting suggests that the firm reduced its holdings to just over 8,000 BTC. This aligns with Wall Street Journal findings in August 2023, indicating that the firm had sold the asset. Similarly, Tesla purchased over 43,000 BTC worth $1.5 billion in February 2021, according to Securities and Exchange Commission filings. The firm cited a need 'for more flexibility to diversify and maximize returns' on its cash as the reason for its decision. Like SpaceX, Tesla has also significantly reduced its Bitcoin holdings to just over 11,500 BTC, worth $1.4 billion at last his firms have significantly reduced their Bitcoin holdings in recent years, Musk in 2022 said that he intended not to sell his holdings. He said this while discussing inflation concerns. 'It is generally better to own physical things like a home or stock in companies you think make good products, than dollars when inflation is high,' he said on X at the time. 'I still own & won't sell my Bitcoin, Ethereum or Doge fwiw.' Recently, Musk has stoked speculation that he has been quietly accumulating the asset by liking a post suggesting so. He has also said his proposed political party, the 'America party,' would 'embrace Bitcoin,' slamming fiat as 'hopeless' in the wake of his disillusionment with the government's decision to pass the so-called "One Big Beautiful Bill." The act's provisions could add over $3 trillion to the national deficit in the next decade, according to estimates by the Congressional Budget Office in May. Read Next: A must-have for all crypto enthusiasts: . Image: Shutterstock This article Elon Musk's SpaceX Moves Bitcoin For The First Time In 3 Years: Is A Sell-Off Incoming? originally appeared on Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data


Forbes
24 minutes ago
- Forbes
AI Alone Won't Save You: Solve A Real Problem First
Digital generated image of abstract AI data chat icons flying over digital surface with codes We're living in the age of copy-paste intelligence. You can spin up an AI-powered landing page in 15 minutes, plug in a few GPT calls, and it'll even write its own press release. The tools are impressive. But most of what gets built doesn't stick. The real issue isn't a lack of AI, it's a lack of purpose. Many products skip the part where they answer: Why does this need to exist? Too many products today are beautifully packaged but hollow. A glossy AI wrapper with no meaningful problem underneath. Founders get caught up in the novelty and forget to ask the most important question: What are we really solving? Great products start at the breaking points. Where something is frustrating, slow, tedious, or emotionally draining. That's where the value is. AI should quietly support that, not be the entire pitch. Start With Problem, Not the Solution A lot of AI products today start with the technology. The model comes first. Then comes the scramble for a use case. That's backwards. The best ideas start with a real-world problem. Something that's annoying, inefficient, or consistently painful. That's where AI can make a real difference. Take something as specific (and surprisingly frustrating) as naming a company. It sounds simple, until you try. Founders spend hours searching for something that feels right - a name that fits the story, the tone, the ambition. But they often settle for whatever's available. Keyword-based search tools don't help much in finding great domain names. You enter a word, get back a bunch of literal matches. Some might be close. Most aren't. But that's not how people think. They search in feelings. In phrases. 'Something calming but confident.' 'A name that sounds smart but not cold.' That's not a search problem. It's a language problem. We built semantic search at not to showcase AI, but to bridge the gap between how people describe what they want and how results are typically delivered. One founder, for example, was building a modern parenting brand. She didn't search for 'baby' or 'mom.' She searched for something that felt timeless and trustworthy, something that could evolve with the brand. The name she picked wasn't obvious. But it clicked. Because it captured the feeling, not just the words. That's what AI should do. Not call attention to itself, but quietly help people get to a better answer. The Real Difference Is in the Decisions Everyone has access to the same tools. The same models, APIs, tutorials. That's not where the edge is. What really separates strong products from forgettable ones is the thinking behind them. The decisions about what to build, and what to leave out. It takes restraint to focus on solving one specific thing well. It takes clarity to ignore the hype and focus on something that's actually useful. The best builders aren't showing off the technology. They're paying attention to the person on the other side. The Best AI Products Don't Talk About AI Think about the tools you rely on every day. Notion. Figma. Superhuman. They all use AI but none of them lead with it. They just work better. Quietly. That's what progress looks like. You don't remember the tech. You remember how much smoother things felt. One of the clearest examples of this approach is Dharmesh Shah, co-founder of HubSpot. He's quietly building a suite of AI-powered tools that actually help you get work done. Agents that research companies, summarize complex inputs, and yes, even help find domain names. Not as a gimmick. But because those are tasks people genuinely want to spend less time on. It's not about being flashy. It's about being helpful. Better Questions to Ask If AI can make that feel lighter, faster, or simpler, you're onto something. If you're starting with 'what can we build with GPT-4,' you're probably building a demo, not a product. Final Thought: Solve Like a Human, Then Scale Like a Machine The companies that will stand out in this next wave won't be the ones shouting about AI. They'll be the ones quietly removing friction. Making things easier. Helping people move faster with less frustration. That's the work that lasts. Start with pain. Then let the AI fade into the background.