Is SFTBY a Buy or a Sell? The Bull and Bear Case (2026)
Last updated October 2026
Short answer
Both cases are real, which is why the question is contested. The bull case for SoftBank Group Corp. (SFTBY) rests on OpenAI valuation and a possible IPO: OpenAI is now SoftBank's single largest swing factor, accounting for most of the roughly $45 billion of cumulative investment gains booked so far. The bear case rests on concentration is the core risk: a large share of NAV now rests on two assets, one of them private and valued at funding-round prices that could be marked down. We do not give a verdict. What follows is each case at full strength, so you can decide which set of assumptions you actually believe. Walnut is not an investment adviser.
SoftBank Group Corp. is a Japanese strategic investment holding company led by founder Masayoshi Son. Its main assets are a roughly 90% stake in chip designer Arm Holdings, a stake of about 13% in OpenAI (built from cumulative commitments of roughly $64.6 billion by October 2026), a large holding in Japanese telecom SoftBank Corp., and the Vision Fund portfolios, which hold stakes in companies such as Coupang, DiDi and Klarna along with a ByteDance position. Over 2025 and 2026 it sold holdings including Nvidia and T-Mobile US to fund its AI commitments, took a stake in Intel, and backed the Stargate data center build-out with OpenAI and Oracle, with Son named chairman of that venture. It has also announced a planned Ohio power and data center campus that a SoftBank executive described as having about 10 gigawatts of planned capacity. In practice, the group has moved from a diversified venture investor toward a concentrated, debt-funded vehicle for AI infrastructure and model companies. The investment picture is about net asset value (NAV) and leverage. SoftBank reported NAV of about ¥72 trillion at June 30, 2026 with a loan-to-value ratio near 13%, but the shares have typically traded at a large discount to that figure, and the price is highly sensitive to OpenAI's private valuation, Arm's share price and news about how SoftBank finances its commitments. Earnings swing with mark-to-market gains: the Vision Fund booked about $46 billion of gains in the year to March 2026, mostly from OpenAI, while the April to June 2026 quarter's ¥347 billion profit leaned on Intel instead. US investors typically hold the company through the SFTBY ADR, which trades over the counter and represents a fraction of one Tokyo-listed share. That means an ADR holder carries three layers of exposure at once: the value of SoftBank's underlying stakes, the size of the discount the market applies to the holding company, and the yen to dollar exchange rate. The ADR's 52-week range of roughly $10.55 to $28.97 in 2026 data listings gives a sense of how far those layers can move together.
The bull case for SFTBY
1. OpenAI valuation and a possible IPO
OpenAI is now SoftBank's single largest swing factor, accounting for most of the roughly $45 billion of cumulative investment gains booked so far. Because OpenAI is private, the stake is marked at funding-round valuations and cannot easily be turned into cash, so reporting about IPO timing (delayed toward 2027 in mid-2026 reports) has moved the stock sharply in both directions. SoftBank committed roughly $30 billion more to OpenAI during 2026, with tranches such as the $10 billion installment scheduled for October 1, so each funding step adds to both the potential gain and the concentration. A public listing would give the stake a market price and a path to liquidity, while a lower private valuation in a future round would flow straight into reported losses.
2. Arm Holdings
SoftBank owns about 90% of Arm, the Nasdaq-listed chip architecture company whose designs sit inside nearly every smartphone and a growing share of data center CPUs. Arm's share price feeds straight into SoftBank's NAV, and Arm shares are also a key source of collateral for SoftBank's margin loans, so a large move in Arm tends to show up in SFTBY within days. Arm's licensing and royalty model gives it high margins, and its push into AI data center chips has been a central part of the bull case for both Arm and SoftBank. In one stretch of 2026, Arm rose more than 30% across two sessions and SoftBank gained close to 25% over five sessions, which shows how tightly the two are linked.
3. Financing capacity and leverage discipline
Funding the OpenAI and Stargate commitments has required asset sales, a reported $40 billion bridge loan, a $10 billion loan secured on the OpenAI stake, and an $11.1 billion high-yield bond sale in September 2026. SoftBank's stated policy is to keep LTV below 25% in normal times, and the market watches that ratio, along with any new borrowing, as a measure of how much room the group has left. The September bond deal was described by Reuters as the largest high-yield corporate bond sale on record, which signals both strong investor demand and how much capital the AI program requires. Lenders were initially hesitant to accept OpenAI shares as collateral because the company is private and hard to value, so the terms of each new facility are themselves a signal of how the market views SoftBank's assets.
4. The NAV discount itself
SoftBank has spent years arguing that its shares trade too far below NAV and has used buybacks to close the gap. Third-party estimates put the discount above 50% in early 2026 and narrower during the mid-2026 rally, so changes in the discount can move the share price as much as changes in the underlying assets. The company's own investor materials attribute the discount to portfolio performance, uncertainty over asset values and the complexity of its structure. A listing of OpenAI, further buybacks or simpler disclosure are the paths most often cited for narrowing it, while new leverage or write-downs tend to widen it.
The bear case for SFTBY
Concentration is the core risk: a large share of NAV now rests on two assets, one of them private and valued at funding-round prices that could be marked down. Leverage magnifies every move, and margin loans secured on Arm and OpenAI shares can force asset sales if those values fall. Key-person risk around Masayoshi Son is unusually high, and the Vision Fund has a record of large write-downs as well as gains. For US holders, SFTBY is an over-the-counter ADR priced in dollars on a yen-denominated stock, so returns include yen moves, OTC liquidity can be thinner than in Tokyo, and the ADR ratio and any depositary fees affect what each ADR represents. There is also execution risk in Stargate and the data center campuses, which require large amounts of capital, power and construction before they generate revenue. Finally, the timing of an OpenAI IPO is outside SoftBank's control, and mid-2026 reports of a delay to 2027 showed how quickly sentiment can reverse when that timeline shifts.
The bear case deserves the same attention as the bull case, and usually gets less. If you are holding SFTBY already, the question is not whether these risks exist but whether any of them has moved from possible to actually happening in the reported numbers.
Where analysts land on SFTBY
Too few analysts publish on SFTBY for a consensus target to mean anything, so there is no professional average to weigh against your own view. That cuts both ways: less informed opinion to lean on, and less of it already priced in. The SFTBY forecast page covers what coverage does exist.
How is SFTBY valued? (as of October 2026)
Snapshot for SFTBY as of October 2026, sourced from Yahoo Finance and may be delayed. Valuation figures move with price and earnings; verify the current numbers with your broker before deciding.
- Net asset value (June 30, 2026): ~¥72 trillion (company figure; ~¥58 trillion in early-August press estimates)
- Loan-to-value ratio: ~13% (policy ceiling 25%)
- Q1 FY2026 net income (Apr to Jun 2026): ~¥347 billion, down ~18% year on year
- Vision Fund gain, FY to March 2026: ~$46 billion, mostly OpenAI
- OpenAI stake: ~13%, ~$64.6 billion cumulative investment
- ADR structure: ~0.5 ordinary share per SFTBY ADR (two ADRs per Tokyo share)
SoftBank is usually valued on NAV minus a discount rather than on earnings, because profits are dominated by mark-to-market gains and losses on its holdings. The company reported NAV per share near ¥12,700 at June 30, 2026 against a Tokyo price near ¥6,000, though press estimates of NAV were lower by August as listed holdings fell, so the discount moves with both the assets and the share price. The 4-for-1 share split effective January 1, 2026 restated all per-share figures. The dividend is small relative to the share price (a post-split forecast of ¥11 per share for the full year to March 2027), so returns for holders have come almost entirely from price changes rather than income.
How do you decide if SFTBY is a buy?
Rather than asking whether SFTBY is a buy in the abstract, it tends to help to answer four questions:
- Thesis: do you believe the bull case above, and is it still true today?
- Time horizon: a single stock can be volatile, so a longer horizon absorbs more of the swings.
- Position sizing: a thesis can be right and the sizing still wrong; decide how much of your portfolio one name should be.
- Overlap: check whether you already hold SFTBY indirectly through an index or sector ETF before adding more.
What would change your mind on SFTBY
Write the tripwires down before you need them. Deciding what would falsify your view is far harder once a position is moving against you.
- Bull case breaks if: OpenAI valuation and a possible IPO stalls in the reported numbers rather than in the narrative around them.
- Bear case breaks if: concentration is the core risk: a large share of NAV now rests on two assets, one of them private and valued at funding-round prices that could be marked down fails to materialise over several reporting periods while the drivers keep compounding.
- Neither matters if: the position has grown large enough that being wrong would damage the whole portfolio. Sizing overrides the argument.
For the full picture, see the SFTBY stock guide (what the company does, the ETFs that hold it, similar stocks, and the themes it fits). In Walnut you can ask its AI about SFTBY against your real portfolio and see your actual exposure before deciding.
Investing in SoftBank Group Corp. with AI
Connect the broker you already use and ask Walnut's AI how SFTBY fits what you actually hold: whether you own it already through a fund, what it would do to your concentration, and how it has tracked the S&P 500. Read-only by default, and you approve anything before it reaches your broker.
FAQ
Is SFTBY a good stock to buy right now?
+
That depends on which case you find more convincing, and both are on this page. The bull case rests on OpenAI valuation and a possible IPO, with net asset value (june 30, 2026) at ~¥72 trillion (company figure; ~¥58 trillion in early-August press estimates). The bear case rests on concentration is the core risk: a large share of NAV now rests on two assets, one of them private and valued at funding-round prices that could be marked down. If you believe the thesis, the real questions become sizing and overlap rather than timing. Walnut is not an investment adviser and this is not a recommendation.
Should I sell SFTBY?
+
Nobody can answer that for you, and the honest version of the question is narrower: has anything changed in the reason you bought it? The bear case on this page is the place to check. Concentration is the core risk: a large share of NAV now rests on two assets, one of them private and valued at funding-round prices that could be marked down. If that risk is what you were worried about and it is now playing out, that is a real signal. If the price simply fell while the thesis held, that is a different situation entirely. Walnut is not an investment adviser.
What is the bull case for SFTBY?
+
OpenAI valuation and a possible IPO. OpenAI is now SoftBank's single largest swing factor, accounting for most of the roughly $45 billion of cumulative investment gains booked so far.
What is the bear case for SFTBY?
+
Concentration is the core risk: a large share of NAV now rests on two assets, one of them private and valued at funding-round prices that could be marked down. Leverage magnifies every move, and margin loans secured on Arm and OpenAI shares can force asset sales if those values fall. Key-person risk around Masayoshi Son is unusually high, and the Vision Fund has a record of large write-downs as well as gains. For US holders, SFTBY is an over-the-counter ADR priced in dollars on a yen-denominated stock, so returns include yen moves, OTC liquidity can be thinner than in Tokyo, and the ADR ratio and any depositary fees affect what each ADR represents. There is also execution risk in Stargate and the data center campuses, which require large amounts of capital, power and construction before they generate revenue. Finally, the timing of an OpenAI IPO is outside SoftBank's control, and mid-2026 reports of a delay to 2027 showed how quickly sentiment can reverse when that timeline shifts.
What does SoftBank Group Corp. do?
+
Tokyo-based investment holding company with controlling stakes in Arm and SoftBank Corp., the Vision Funds and a large OpenAI stake.
What would have to change for SFTBY to stop being worth holding?
+
Decide that in advance, because it is far harder to think clearly once a position is moving against you. The concrete tripwires here: the driver behind the bull case (OpenAI valuation and a possible IPO) stalling in the reported numbers rather than in the narrative, the risk above (concentration is the core risk: a large share of NAV now rests on two assets, one of them private and valued at funding-round prices that could be marked down) turning from a possibility into a reported fact, or the position growing large enough that a bad outcome would matter to your whole portfolio regardless of who is right.
What is SFTBY?
+
SFTBY is the ticker for the American depositary receipt of SoftBank Group Corp., which is listed in Tokyo under the code 9984. The ADR trades over the counter in the US rather than on the NYSE or Nasdaq, and it lets US investors hold the stock in dollars through an ordinary brokerage account. Each SFTBY ADR represents a fraction of one Tokyo share (about half a share, or two ADRs per ordinary share), and its price tracks the Tokyo price converted at the prevailing yen to dollar exchange rate. Because the ADR is a claim on Tokyo shares held by a depositary bank, its value is anchored to the Tokyo market, but prices can drift briefly when Tokyo is closed and US trading reacts to news first.
What does SoftBank Group actually own?
+
SoftBank Group is a holding company rather than an operating business. Its largest assets are about 90% of Arm Holdings, a stake of roughly 13% in OpenAI, a controlling stake in Japanese mobile carrier SoftBank Corp., and the two Vision Funds, which hold stakes in companies such as Coupang, DiDi, Klarna and ByteDance. It has also taken a stake in Intel. Over 2025 and 2026 it sold positions including Nvidia and T-Mobile US to raise cash for its AI commitments, so the portfolio has become increasingly concentrated in AI. The Intel stake contributed about ¥1.86 trillion of investment gains in the April to June 2026 quarter, which shows that even smaller positions can move results. SoftBank Corp., the telecom unit, is the most stable cash-generating holding, but it is a much smaller share of NAV than Arm and OpenAI.
How big is SoftBank's OpenAI stake?
+
Reports in August and September 2026 put SoftBank's cumulative investment in OpenAI at about $64.6 billion by October 2026, for a stake of roughly 13%. The Vision Fund booked about $46 billion of gains for the fiscal year to March 2026, mostly from OpenAI, and SoftBank has cited cumulative OpenAI gains near $45 billion. Because OpenAI is private, the stake is valued at funding-round prices, and it cannot be sold quickly. SoftBank has borrowed against it, including a $10 billion loan arranged in August 2026. SoftBank also arranged a reported $40 billion bridge loan to cover its 2026 commitments, and an OpenAI IPO, if it happens, would be the most direct way for that stake to become liquid.
Walnut is informational, not investment advice, and gives no verdict on SFTBY. Analyst targets referenced here come from a October 2026 pull of published third-party research and change constantly. Verify current figures with your broker before acting on them.